2_finetuning.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/olachinkeigpu/anaconda3/envs/llm-finetuning/lib/python3.11/site-packages/tqdm/auto.py:21: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html\n",
      "  from .autonotebook import tqdm as notebook_tqdm\n"
     ]
    }
   ],
   "source": [
    "import bitsandbytes\n",
    "import os\n",
    "import wandb\n",
    "import json\n",
    "import tqdm\n",
    "from datasets import Dataset, concatenate_datasets, load_dataset\n",
    "import transformers\n",
    "from transformers import AutoTokenizer, AutoModelForCausalLM\n",
    "import torch\n",
    "import torch.nn as nn\n",
    "from peft import LoraConfig\n",
    "import wandb\n",
    "from trl import DataCollatorForCompletionOnlyLM, SFTTrainer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/bin/bash: -c: line 2: syntax error: unexpected end of file\n"
     ]
    }
   ],
   "source": [
    "wandb.login()\n",
    "os.environ[\"CUDA_VISIBLE_DEVICES\"]=\"0\"\n",
    "os.environ[\"WANDB_LOG_MODEL\"] = \"checkpoint\"\n",
    "os.environ[\"WANDB_WATCH\"] = \"gradients\"\n",
    "os.environ[\"PYTORCH_CUDA_ALLOC_CONF\"] = \"max_split_size_mb=1024\"\n",
    "os.environ[\"TOKENIZERS_PARALLELISM\"] = \"false\"\n",
    "wandb_entity=\"\" # change here\n",
    "wandb_project=\"\" # change here"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "config = {\n",
    "    \"BASE_MODEL\":\"cyberagent/open-calm-medium\",\n",
    "    \"model_artifacts\": '', # change here\n",
    "    \"tuning_data_artifacts\": {\"Anthropic_hh_rlfh\": '', # change here\n",
    "                              \"OpenAssistant_oasst1\": '', # change here\n",
    "                              \"databricks-dolly-15k-ja\": ''}, # change here\n",
    "    \"max_seq_length\":1024,\n",
    "    \"lora_config\":{\n",
    "        \"r\":8,\n",
    "        \"lora_alpha\":16,\n",
    "        \"target_modules\":[\"query_key_value\"],\n",
    "        \"lora_dropout\":.1,\n",
    "        \"bias\":\"none\",\n",
    "        \"task_type\":\"CAUSAL_LM\"\n",
    "    },\n",
    "    \"training_args\":{\n",
    "        \"dataloader_num_workers\":2,\n",
    "        \"evaluation_strategy\":\"steps\",\n",
    "        \"per_device_train_batch_size\":8,\n",
    "        \"gradient_accumulation_steps\":2,\n",
    "        \"warmup_ratio\": 0.1, \n",
    "        \"warmup_steps\":5,\n",
    "        #\"num_train_epochs\":1,\n",
    "        \"max_steps\": 50,\n",
    "        \"learning_rate\":1e-4,\n",
    "        \"fp16\":True,\n",
    "        \"logging_steps\":10,\n",
    "        \"save_steps\":10,\n",
    "        \"output_dir\":'./outputs',\n",
    "        \"report_to\":\"wandb\"\n",
    "    }\n",
    "}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Failed to detect the name of this notebook, you can set it manually with the WANDB_NOTEBOOK_NAME environment variable to enable code saving.\n",
      "\u001b[34m\u001b[1mwandb\u001b[0m: Currently logged in as: \u001b[33mkeisuke-kamata\u001b[0m (\u001b[33mwandb-japan\u001b[0m). Use \u001b[1m`wandb login --relogin`\u001b[0m to force relogin\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "Tracking run with wandb version 0.16.2"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "Run data is saved locally in <code>/home/olachinkeigpu/Project/llm-finetuning-handson/wandb/run-20240123_102542-3e15qgaf</code>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       "Syncing run <strong><a href='https://wandb.ai/wandb-japan/llm-finetuning/runs/3e15qgaf' target=\"_blank\">confused-wind-156</a></strong> to <a href='https://wandb.ai/wandb-japan/llm-finetuning' target=\"_blank\">Weights & Biases</a> (<a href='https://wandb.me/run' target=\"_blank\">docs</a>)<br/>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       " View project at <a href='https://wandb.ai/wandb-japan/llm-finetuning' target=\"_blank\">https://wandb.ai/wandb-japan/llm-finetuning</a>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/html": [
       " View run at <a href='https://wandb.ai/wandb-japan/llm-finetuning/runs/3e15qgaf' target=\"_blank\">https://wandb.ai/wandb-japan/llm-finetuning/runs/3e15qgaf</a>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "\u001b[34m\u001b[1mwandb\u001b[0m: Downloading large artifact cyberagent-open-calm-medium:v0, 783.23MB. 6 files... \n",
      "\u001b[34m\u001b[1mwandb\u001b[0m:   6 of 6 files downloaded.  \n",
      "Done. 0:0:1.0\n",
      "\u001b[34m\u001b[1mwandb\u001b[0m: Downloading large artifact Anthropic_hh_rlfh:v0, 68.05MB. 1 files... \n",
      "\u001b[34m\u001b[1mwandb\u001b[0m:   1 of 1 files downloaded.  \n",
      "Done. 0:0:0.3\n",
      "\u001b[34m\u001b[1mwandb\u001b[0m: Downloading large artifact OpenAssistant_oasst1:v0, 114.69MB. 1 files... \n",
      "\u001b[34m\u001b[1mwandb\u001b[0m:   1 of 1 files downloaded.  \n",
      "Done. 0:0:0.4\n",
      "\u001b[34m\u001b[1mwandb\u001b[0m:   1 of 1 files downloaded.  \n",
      "/home/olachinkeigpu/anaconda3/envs/llm-finetuning/lib/python3.11/site-packages/trl/trainer/utils.py:104: UserWarning: The pad_token_id and eos_token_id values of this tokenizer are identical. If you are planning for multi-turn training, it can result in the model continuously generating questions and answers without eos token. To avoid this, set the pad_token_id to a different value.\n",
      "  warnings.warn(\n",
      "Map: 100%|██████████| 102592/102592 [00:19<00:00, 5163.97 examples/s]\n",
      "Map: 100%|██████████| 11400/11400 [00:02<00:00, 5059.16 examples/s]\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "/home/olachinkeigpu/anaconda3/envs/llm-finetuning/lib/python3.11/site-packages/torch/utils/checkpoint.py:429: UserWarning: torch.utils.checkpoint: please pass in use_reentrant=True or use_reentrant=False explicitly. The default value of use_reentrant will be updated to be False in the future. To maintain current behavior, pass use_reentrant=True. It is recommended that you use use_reentrant=False. Refer to docs for more details on the differences between the two variants.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "\n",
       "    <div>\n",
       "      \n",
       "      <progress value='2631' max='6412' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      [2631/6412 4:05:12 < 5:52:39, 0.18 it/s, Epoch 0.41/1]\n",
       "    </div>\n",
       "    <table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       " <tr style=\"text-align: left;\">\n",
       "      <th>Step</th>\n",
       "      <th>Training Loss</th>\n",
       "      <th>Validation Loss</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>10</td>\n",
       "      <td>3.515400</td>\n",
       "      <td>3.437527</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>20</td>\n",
       "      <td>3.538800</td>\n",
       "      <td>3.410272</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>30</td>\n",
       "      <td>3.421000</td>\n",
       "      <td>3.389726</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>40</td>\n",
       "      <td>3.239700</td>\n",
       "      <td>3.377525</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>50</td>\n",
       "      <td>3.396600</td>\n",
       "      <td>3.369113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>60</td>\n",
       "      <td>3.481700</td>\n",
       "      <td>3.362709</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>70</td>\n",
       "      <td>3.230200</td>\n",
       "      <td>3.357495</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>80</td>\n",
       "      <td>3.252100</td>\n",
       "      <td>3.352502</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>90</td>\n",
       "      <td>3.390300</td>\n",
       "      <td>3.348202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>100</td>\n",
       "      <td>3.383600</td>\n",
       "      <td>3.344912</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>110</td>\n",
       "      <td>3.467900</td>\n",
       "      <td>3.342193</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>120</td>\n",
       "      <td>3.380000</td>\n",
       "      <td>3.339278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>130</td>\n",
       "      <td>3.374400</td>\n",
       "      <td>3.336942</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>140</td>\n",
       "      <td>3.141500</td>\n",
       "      <td>3.334818</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>150</td>\n",
       "      <td>3.425700</td>\n",
       "      <td>3.332937</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>160</td>\n",
       "      <td>3.335300</td>\n",
       "      <td>3.330999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>170</td>\n",
       "      <td>3.460100</td>\n",
       "      <td>3.329473</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>180</td>\n",
       "      <td>3.353000</td>\n",
       "      <td>3.327725</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>190</td>\n",
       "      <td>3.307800</td>\n",
       "      <td>3.326198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>200</td>\n",
       "      <td>3.312800</td>\n",
       "      <td>3.324839</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>210</td>\n",
       "      <td>3.407000</td>\n",
       "      <td>3.323555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>220</td>\n",
       "      <td>3.281700</td>\n",
       "      <td>3.322726</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>230</td>\n",
       "      <td>3.235300</td>\n",
       "      <td>3.321635</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>240</td>\n",
       "      <td>3.378200</td>\n",
       "      <td>3.320873</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>250</td>\n",
       "      <td>3.395400</td>\n",
       "      <td>3.320021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>260</td>\n",
       "      <td>3.367700</td>\n",
       "      <td>3.319287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>270</td>\n",
       "      <td>3.323400</td>\n",
       "      <td>3.318244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>280</td>\n",
       "      <td>3.272000</td>\n",
       "      <td>3.317488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>290</td>\n",
       "      <td>3.349100</td>\n",
       "      <td>3.316956</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>300</td>\n",
       "      <td>3.320000</td>\n",
       "      <td>3.316419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>310</td>\n",
       "      <td>3.435900</td>\n",
       "      <td>3.315551</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>320</td>\n",
       "      <td>3.337300</td>\n",
       "      <td>3.314750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>330</td>\n",
       "      <td>3.282900</td>\n",
       "      <td>3.314177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>340</td>\n",
       "      <td>3.434300</td>\n",
       "      <td>3.313341</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>350</td>\n",
       "      <td>3.289000</td>\n",
       "      <td>3.312610</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>360</td>\n",
       "      <td>3.338500</td>\n",
       "      <td>3.311696</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>370</td>\n",
       "      <td>3.305000</td>\n",
       "      <td>3.311207</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>380</td>\n",
       "      <td>3.311800</td>\n",
       "      <td>3.310915</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>390</td>\n",
       "      <td>3.292100</td>\n",
       "      <td>3.310475</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>400</td>\n",
       "      <td>3.257600</td>\n",
       "      <td>3.310017</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>410</td>\n",
       "      <td>3.331900</td>\n",
       "      <td>3.309424</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>420</td>\n",
       "      <td>3.319900</td>\n",
       "      <td>3.308491</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>430</td>\n",
       "      <td>3.322500</td>\n",
       "      <td>3.308586</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>440</td>\n",
       "      <td>3.315400</td>\n",
       "      <td>3.308248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>450</td>\n",
       "      <td>3.354900</td>\n",
       "      <td>3.307243</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>460</td>\n",
       "      <td>3.354400</td>\n",
       "      <td>3.306809</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>470</td>\n",
       "      <td>3.181800</td>\n",
       "      <td>3.306615</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>480</td>\n",
       "      <td>3.346300</td>\n",
       "      <td>3.306324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>490</td>\n",
       "      <td>3.373500</td>\n",
       "      <td>3.305852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>500</td>\n",
       "      <td>3.320600</td>\n",
       "      <td>3.305066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>510</td>\n",
       "      <td>3.313200</td>\n",
       "      <td>3.304685</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>520</td>\n",
       "      <td>3.254700</td>\n",
       "      <td>3.304744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>530</td>\n",
       "      <td>3.307400</td>\n",
       "      <td>3.304202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>540</td>\n",
       "      <td>3.160800</td>\n",
       "      <td>3.303585</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>550</td>\n",
       "      <td>3.320900</td>\n",
       "      <td>3.303565</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>560</td>\n",
       "      <td>3.227200</td>\n",
       "      <td>3.303474</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>570</td>\n",
       "      <td>3.266500</td>\n",
       "      <td>3.302772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>580</td>\n",
       "      <td>3.306400</td>\n",
       "      <td>3.302790</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>590</td>\n",
       "      <td>3.355400</td>\n",
       "      <td>3.302147</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>600</td>\n",
       "      <td>3.399800</td>\n",
       "      <td>3.301946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>610</td>\n",
       "      <td>3.381400</td>\n",
       "      <td>3.301421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>620</td>\n",
       "      <td>3.307400</td>\n",
       "      <td>3.300722</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>630</td>\n",
       "      <td>3.276400</td>\n",
       "      <td>3.300083</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>640</td>\n",
       "      <td>3.340100</td>\n",
       "      <td>3.300190</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>650</td>\n",
       "      <td>3.285900</td>\n",
       "      <td>3.299888</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>660</td>\n",
       "      <td>3.351400</td>\n",
       "      <td>3.299185</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>670</td>\n",
       "      <td>3.248200</td>\n",
       "      <td>3.299298</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>680</td>\n",
       "      <td>3.203600</td>\n",
       "      <td>3.298786</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>690</td>\n",
       "      <td>3.346400</td>\n",
       "      <td>3.298418</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>700</td>\n",
       "      <td>3.327800</td>\n",
       "      <td>3.297871</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>710</td>\n",
       "      <td>3.246400</td>\n",
       "      <td>3.297485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>720</td>\n",
       "      <td>3.373500</td>\n",
       "      <td>3.297693</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>730</td>\n",
       "      <td>3.257100</td>\n",
       "      <td>3.296978</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>740</td>\n",
       "      <td>3.219700</td>\n",
       "      <td>3.296595</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>750</td>\n",
       "      <td>3.254200</td>\n",
       "      <td>3.296075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>760</td>\n",
       "      <td>3.191900</td>\n",
       "      <td>3.295887</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>770</td>\n",
       "      <td>3.357900</td>\n",
       "      <td>3.296068</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>780</td>\n",
       "      <td>3.279200</td>\n",
       "      <td>3.295562</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>790</td>\n",
       "      <td>3.254200</td>\n",
       "      <td>3.295979</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>800</td>\n",
       "      <td>3.370900</td>\n",
       "      <td>3.296130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>810</td>\n",
       "      <td>3.238000</td>\n",
       "      <td>3.295027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>820</td>\n",
       "      <td>3.358900</td>\n",
       "      <td>3.294513</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>830</td>\n",
       "      <td>3.270400</td>\n",
       "      <td>3.294446</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>840</td>\n",
       "      <td>3.427100</td>\n",
       "      <td>3.294546</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>850</td>\n",
       "      <td>3.314600</td>\n",
       "      <td>3.294036</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>860</td>\n",
       "      <td>3.207300</td>\n",
       "      <td>3.294142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>870</td>\n",
       "      <td>3.321400</td>\n",
       "      <td>3.293480</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>880</td>\n",
       "      <td>3.178500</td>\n",
       "      <td>3.293190</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>890</td>\n",
       "      <td>3.327800</td>\n",
       "      <td>3.292612</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>900</td>\n",
       "      <td>3.269900</td>\n",
       "      <td>3.292401</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>910</td>\n",
       "      <td>3.221200</td>\n",
       "      <td>3.292130</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>920</td>\n",
       "      <td>3.332600</td>\n",
       "      <td>3.291975</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>930</td>\n",
       "      <td>3.405100</td>\n",
       "      <td>3.292097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>940</td>\n",
       "      <td>3.333600</td>\n",
       "      <td>3.291404</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>950</td>\n",
       "      <td>3.309100</td>\n",
       "      <td>3.291129</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>960</td>\n",
       "      <td>3.345400</td>\n",
       "      <td>3.290918</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>970</td>\n",
       "      <td>3.315800</td>\n",
       "      <td>3.291393</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>980</td>\n",
       "      <td>3.259900</td>\n",
       "      <td>3.290486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>990</td>\n",
       "      <td>3.278600</td>\n",
       "      <td>3.290402</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1000</td>\n",
       "      <td>3.360900</td>\n",
       "      <td>3.289878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1010</td>\n",
       "      <td>3.273100</td>\n",
       "      <td>3.289631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1020</td>\n",
       "      <td>3.309400</td>\n",
       "      <td>3.289497</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1030</td>\n",
       "      <td>3.290500</td>\n",
       "      <td>3.289211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1040</td>\n",
       "      <td>3.287500</td>\n",
       "      <td>3.288667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1050</td>\n",
       "      <td>3.394100</td>\n",
       "      <td>3.288833</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1060</td>\n",
       "      <td>3.348600</td>\n",
       "      <td>3.288792</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1070</td>\n",
       "      <td>3.340300</td>\n",
       "      <td>3.288458</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1080</td>\n",
       "      <td>3.254000</td>\n",
       "      <td>3.287843</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1090</td>\n",
       "      <td>3.427400</td>\n",
       "      <td>3.288481</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1100</td>\n",
       "      <td>3.352100</td>\n",
       "      <td>3.288094</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1110</td>\n",
       "      <td>3.294400</td>\n",
       "      <td>3.287384</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1120</td>\n",
       "      <td>3.337500</td>\n",
       "      <td>3.287428</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1130</td>\n",
       "      <td>3.298800</td>\n",
       "      <td>3.287856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1140</td>\n",
       "      <td>3.337400</td>\n",
       "      <td>3.287293</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1150</td>\n",
       "      <td>3.268200</td>\n",
       "      <td>3.286625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1160</td>\n",
       "      <td>3.344500</td>\n",
       "      <td>3.287276</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1170</td>\n",
       "      <td>3.361600</td>\n",
       "      <td>3.286435</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1180</td>\n",
       "      <td>3.307700</td>\n",
       "      <td>3.285972</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1190</td>\n",
       "      <td>3.265500</td>\n",
       "      <td>3.285835</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1200</td>\n",
       "      <td>3.331600</td>\n",
       "      <td>3.285639</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1210</td>\n",
       "      <td>3.276600</td>\n",
       "      <td>3.285703</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1220</td>\n",
       "      <td>3.394700</td>\n",
       "      <td>3.285652</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1230</td>\n",
       "      <td>3.249100</td>\n",
       "      <td>3.285582</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1240</td>\n",
       "      <td>3.298900</td>\n",
       "      <td>3.285194</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1250</td>\n",
       "      <td>3.195300</td>\n",
       "      <td>3.284772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1260</td>\n",
       "      <td>3.280100</td>\n",
       "      <td>3.284991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1270</td>\n",
       "      <td>3.243600</td>\n",
       "      <td>3.284553</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1280</td>\n",
       "      <td>3.227700</td>\n",
       "      <td>3.284354</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1290</td>\n",
       "      <td>3.254400</td>\n",
       "      <td>3.284319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1300</td>\n",
       "      <td>3.236100</td>\n",
       "      <td>3.284338</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1310</td>\n",
       "      <td>3.395000</td>\n",
       "      <td>3.283555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1320</td>\n",
       "      <td>3.211400</td>\n",
       "      <td>3.283868</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1330</td>\n",
       "      <td>3.355100</td>\n",
       "      <td>3.283517</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1340</td>\n",
       "      <td>3.337000</td>\n",
       "      <td>3.283513</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1350</td>\n",
       "      <td>3.266500</td>\n",
       "      <td>3.283793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1360</td>\n",
       "      <td>3.291500</td>\n",
       "      <td>3.283172</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1370</td>\n",
       "      <td>3.340000</td>\n",
       "      <td>3.283290</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1380</td>\n",
       "      <td>3.270100</td>\n",
       "      <td>3.283259</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1390</td>\n",
       "      <td>3.202700</td>\n",
       "      <td>3.282637</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1400</td>\n",
       "      <td>3.300600</td>\n",
       "      <td>3.282257</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1410</td>\n",
       "      <td>3.393500</td>\n",
       "      <td>3.282066</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1420</td>\n",
       "      <td>3.167300</td>\n",
       "      <td>3.281780</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1430</td>\n",
       "      <td>3.214600</td>\n",
       "      <td>3.281845</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1440</td>\n",
       "      <td>3.353800</td>\n",
       "      <td>3.281604</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1450</td>\n",
       "      <td>3.371300</td>\n",
       "      <td>3.281819</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1460</td>\n",
       "      <td>3.409200</td>\n",
       "      <td>3.281693</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1470</td>\n",
       "      <td>3.209400</td>\n",
       "      <td>3.281237</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1480</td>\n",
       "      <td>3.200500</td>\n",
       "      <td>3.281129</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1490</td>\n",
       "      <td>3.328500</td>\n",
       "      <td>3.281317</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1500</td>\n",
       "      <td>3.164600</td>\n",
       "      <td>3.281168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1510</td>\n",
       "      <td>3.276200</td>\n",
       "      <td>3.280895</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1520</td>\n",
       "      <td>3.277000</td>\n",
       "      <td>3.280733</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1530</td>\n",
       "      <td>3.244700</td>\n",
       "      <td>3.280713</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1540</td>\n",
       "      <td>3.294400</td>\n",
       "      <td>3.279912</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1550</td>\n",
       "      <td>3.274900</td>\n",
       "      <td>3.279863</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1560</td>\n",
       "      <td>3.305300</td>\n",
       "      <td>3.279867</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1570</td>\n",
       "      <td>3.356600</td>\n",
       "      <td>3.280122</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1580</td>\n",
       "      <td>3.317400</td>\n",
       "      <td>3.280097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1590</td>\n",
       "      <td>3.256300</td>\n",
       "      <td>3.279596</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1600</td>\n",
       "      <td>3.287600</td>\n",
       "      <td>3.279372</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1610</td>\n",
       "      <td>3.257500</td>\n",
       "      <td>3.279109</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1620</td>\n",
       "      <td>3.301700</td>\n",
       "      <td>3.279106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1630</td>\n",
       "      <td>3.296700</td>\n",
       "      <td>3.279125</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1640</td>\n",
       "      <td>3.342700</td>\n",
       "      <td>3.279338</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1650</td>\n",
       "      <td>3.177200</td>\n",
       "      <td>3.278775</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1660</td>\n",
       "      <td>3.266200</td>\n",
       "      <td>3.278281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1670</td>\n",
       "      <td>3.249000</td>\n",
       "      <td>3.278612</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1680</td>\n",
       "      <td>3.341200</td>\n",
       "      <td>3.278462</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1690</td>\n",
       "      <td>3.316900</td>\n",
       "      <td>3.278711</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1700</td>\n",
       "      <td>3.285500</td>\n",
       "      <td>3.278033</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1710</td>\n",
       "      <td>3.423800</td>\n",
       "      <td>3.277677</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1720</td>\n",
       "      <td>3.273500</td>\n",
       "      <td>3.278383</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1730</td>\n",
       "      <td>3.372800</td>\n",
       "      <td>3.277987</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1740</td>\n",
       "      <td>3.170700</td>\n",
       "      <td>3.277456</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1750</td>\n",
       "      <td>3.266900</td>\n",
       "      <td>3.277747</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1760</td>\n",
       "      <td>3.258200</td>\n",
       "      <td>3.277318</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1770</td>\n",
       "      <td>3.375700</td>\n",
       "      <td>3.277738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1780</td>\n",
       "      <td>3.358300</td>\n",
       "      <td>3.277623</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1790</td>\n",
       "      <td>3.190100</td>\n",
       "      <td>3.277039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1800</td>\n",
       "      <td>3.163900</td>\n",
       "      <td>3.276898</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1810</td>\n",
       "      <td>3.318000</td>\n",
       "      <td>3.276796</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1820</td>\n",
       "      <td>3.382400</td>\n",
       "      <td>3.276498</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1830</td>\n",
       "      <td>3.246200</td>\n",
       "      <td>3.276491</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1840</td>\n",
       "      <td>3.118800</td>\n",
       "      <td>3.276808</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1850</td>\n",
       "      <td>3.328100</td>\n",
       "      <td>3.276212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1860</td>\n",
       "      <td>3.235300</td>\n",
       "      <td>3.276086</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1870</td>\n",
       "      <td>3.462600</td>\n",
       "      <td>3.276620</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1880</td>\n",
       "      <td>3.357700</td>\n",
       "      <td>3.276097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1890</td>\n",
       "      <td>3.262700</td>\n",
       "      <td>3.275817</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1900</td>\n",
       "      <td>3.220600</td>\n",
       "      <td>3.275753</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1910</td>\n",
       "      <td>3.282400</td>\n",
       "      <td>3.275727</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1920</td>\n",
       "      <td>3.169300</td>\n",
       "      <td>3.275879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1930</td>\n",
       "      <td>3.202200</td>\n",
       "      <td>3.275621</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1940</td>\n",
       "      <td>3.237100</td>\n",
       "      <td>3.275457</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1950</td>\n",
       "      <td>3.161700</td>\n",
       "      <td>3.275280</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1960</td>\n",
       "      <td>3.321400</td>\n",
       "      <td>3.275652</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1970</td>\n",
       "      <td>3.263900</td>\n",
       "      <td>3.275535</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1980</td>\n",
       "      <td>3.217800</td>\n",
       "      <td>3.275203</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1990</td>\n",
       "      <td>3.317400</td>\n",
       "      <td>3.274963</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2000</td>\n",
       "      <td>3.394900</td>\n",
       "      <td>3.274999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2010</td>\n",
       "      <td>3.370500</td>\n",
       "      <td>3.275245</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2020</td>\n",
       "      <td>3.114100</td>\n",
       "      <td>3.274750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2030</td>\n",
       "      <td>3.275300</td>\n",
       "      <td>3.274277</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2040</td>\n",
       "      <td>3.375300</td>\n",
       "      <td>3.274364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2050</td>\n",
       "      <td>3.240100</td>\n",
       "      <td>3.274219</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2060</td>\n",
       "      <td>3.265500</td>\n",
       "      <td>3.273759</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2070</td>\n",
       "      <td>3.316200</td>\n",
       "      <td>3.273682</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2080</td>\n",
       "      <td>3.305500</td>\n",
       "      <td>3.274209</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2090</td>\n",
       "      <td>3.302600</td>\n",
       "      <td>3.273741</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2100</td>\n",
       "      <td>3.321500</td>\n",
       "      <td>3.273398</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2110</td>\n",
       "      <td>3.348400</td>\n",
       "      <td>3.273229</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2120</td>\n",
       "      <td>3.248500</td>\n",
       "      <td>3.273292</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2130</td>\n",
       "      <td>3.229100</td>\n",
       "      <td>3.273368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2140</td>\n",
       "      <td>3.200800</td>\n",
       "      <td>3.273407</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2150</td>\n",
       "      <td>3.241800</td>\n",
       "      <td>3.273304</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2160</td>\n",
       "      <td>3.284200</td>\n",
       "      <td>3.273296</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2170</td>\n",
       "      <td>3.201200</td>\n",
       "      <td>3.273623</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2180</td>\n",
       "      <td>3.392900</td>\n",
       "      <td>3.273292</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2190</td>\n",
       "      <td>3.309100</td>\n",
       "      <td>3.273426</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2200</td>\n",
       "      <td>3.279500</td>\n",
       "      <td>3.272896</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2210</td>\n",
       "      <td>3.278200</td>\n",
       "      <td>3.272777</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2220</td>\n",
       "      <td>3.286200</td>\n",
       "      <td>3.272381</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2230</td>\n",
       "      <td>3.258500</td>\n",
       "      <td>3.272556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2240</td>\n",
       "      <td>3.289900</td>\n",
       "      <td>3.272647</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2250</td>\n",
       "      <td>3.201700</td>\n",
       "      <td>3.272704</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2260</td>\n",
       "      <td>3.225600</td>\n",
       "      <td>3.272339</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2270</td>\n",
       "      <td>3.298600</td>\n",
       "      <td>3.272495</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2280</td>\n",
       "      <td>3.195600</td>\n",
       "      <td>3.272192</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2290</td>\n",
       "      <td>3.327200</td>\n",
       "      <td>3.271738</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2300</td>\n",
       "      <td>3.218700</td>\n",
       "      <td>3.271266</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2310</td>\n",
       "      <td>3.227800</td>\n",
       "      <td>3.271890</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2320</td>\n",
       "      <td>3.377500</td>\n",
       "      <td>3.271111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2330</td>\n",
       "      <td>3.235000</td>\n",
       "      <td>3.271429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2340</td>\n",
       "      <td>3.337400</td>\n",
       "      <td>3.271327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2350</td>\n",
       "      <td>3.293100</td>\n",
       "      <td>3.271441</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2360</td>\n",
       "      <td>3.178700</td>\n",
       "      <td>3.271131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2370</td>\n",
       "      <td>3.379300</td>\n",
       "      <td>3.271082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2380</td>\n",
       "      <td>3.328000</td>\n",
       "      <td>3.271134</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2390</td>\n",
       "      <td>3.277700</td>\n",
       "      <td>3.271286</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2400</td>\n",
       "      <td>3.417200</td>\n",
       "      <td>3.270782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2410</td>\n",
       "      <td>3.286700</td>\n",
       "      <td>3.270669</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2420</td>\n",
       "      <td>3.333900</td>\n",
       "      <td>3.270500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2430</td>\n",
       "      <td>3.090100</td>\n",
       "      <td>3.271081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2440</td>\n",
       "      <td>3.271100</td>\n",
       "      <td>3.270908</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2450</td>\n",
       "      <td>3.307100</td>\n",
       "      <td>3.270204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2460</td>\n",
       "      <td>3.238900</td>\n",
       "      <td>3.270201</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2470</td>\n",
       "      <td>3.293000</td>\n",
       "      <td>3.269952</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2480</td>\n",
       "      <td>3.214700</td>\n",
       "      <td>3.269986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2490</td>\n",
       "      <td>3.286700</td>\n",
       "      <td>3.270097</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2500</td>\n",
       "      <td>3.297600</td>\n",
       "      <td>3.269892</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2510</td>\n",
       "      <td>3.351700</td>\n",
       "      <td>3.270182</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2520</td>\n",
       "      <td>3.247000</td>\n",
       "      <td>3.269856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2530</td>\n",
       "      <td>3.332500</td>\n",
       "      <td>3.270082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2540</td>\n",
       "      <td>3.290000</td>\n",
       "      <td>3.270071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2550</td>\n",
       "      <td>3.270100</td>\n",
       "      <td>3.269450</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2560</td>\n",
       "      <td>3.318200</td>\n",
       "      <td>3.269618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2570</td>\n",
       "      <td>3.336900</td>\n",
       "      <td>3.270399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2580</td>\n",
       "      <td>3.222800</td>\n",
       "      <td>3.270212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2590</td>\n",
       "      <td>3.174000</td>\n",
       "      <td>3.269416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2600</td>\n",
       "      <td>3.192900</td>\n",
       "      <td>3.269289</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2610</td>\n",
       "      <td>3.343200</td>\n",
       "      <td>3.268939</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2620</td>\n",
       "      <td>3.237100</td>\n",
       "      <td>3.268946</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table><p>\n",
       "    <div>\n",
       "      \n",
       "      <progress value='372' max='1425' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
       "      [ 372/1425 00:13 < 00:38, 27.37 it/s]\n",
       "    </div>\n",
       "    "
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "\u001b[34m\u001b[1mwandb\u001b[0m: Adding directory to artifact (./outputs/checkpoint-1000)... Done. 0.0s\n",
      "/home/olachinkeigpu/anaconda3/envs/llm-finetuning/lib/python3.11/site-packages/torch/utils/checkpoint.py:429: UserWarning: torch.utils.checkpoint: please pass in use_reentrant=True or use_reentrant=False explicitly. The default value of use_reentrant will be updated to be False in the future. To maintain current behavior, pass use_reentrant=True. It is recommended that you use use_reentrant=False. Refer to docs for more details on the differences between the two variants.\n",
      "  warnings.warn(\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "\u001b[34m\u001b[1mwandb\u001b[0m: Adding directory to artifact (./outputs/checkpoint-2000)... Done. 0.0s\n",
      "/home/olachinkeigpu/anaconda3/envs/llm-finetuning/lib/python3.11/site-packages/torch/utils/checkpoint.py:429: UserWarning: torch.utils.checkpoint: please pass in use_reentrant=True or use_reentrant=False explicitly. The default value of use_reentrant will be updated to be False in the future. To maintain current behavior, pass use_reentrant=True. It is recommended that you use use_reentrant=False. Refer to docs for more details on the differences between the two variants.\n",
      "  warnings.warn(\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n",
      "You're using a GPTNeoXTokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.\n"
     ]
    }
   ],
   "source": [
    "with wandb.init(entity=wandb_entity, project=wandb_project,  config=config, job_type='finetuning') as run:\n",
    "    model_name = run.config[\"BASE_MODEL\"]\n",
    "    artifact = run.use_artifact(run.config[\"model_artifacts\"], type='model')\n",
    "    model_dir = artifact.download()\n",
    "    torch.cuda.empty_cache()\n",
    "    tokenizer = AutoTokenizer.from_pretrained(model_dir,use_fast=True,trust_remote_code=True)\n",
    "    tokenizer.pad_token = tokenizer.eos_token\n",
    "    tokenizer.pad_token_id = tokenizer.eos_token_id\n",
    "    model = AutoModelForCausalLM.from_pretrained(model_dir, device_map=\"auto\", use_cache=False)\n",
    "\n",
    "    lora_config =  LoraConfig(**wandb.config[\"lora_config\"])\n",
    "\n",
    "\n",
    "    # cast the small parameters (e.g. layernorm) to fp32 for stability\n",
    "    for param in model.parameters():\n",
    "        param.requires_grad = False # freeze the model - train adapters later\n",
    "        if param.ndim == 1:\n",
    "            param.data = param.data.to(torch.float32)\n",
    "    model.gradient_checkpointing_enable()  # reduce number of stored activations\n",
    "    model.enable_input_require_grads()\n",
    "    class CastOutputToFloat(nn.Sequential):\n",
    "        def forward(self, x): return super().forward(x).to(torch.float32)\n",
    "    model.embed_out = CastOutputToFloat(model.embed_out)\n",
    "\n",
    "    datasets = []\n",
    "    # load Anthropic_hh_rlfh\n",
    "    dataset_artifacts=run.use_artifact(run.config[\"tuning_data_artifacts\"][\"Anthropic_hh_rlfh\"], type='dataset')\n",
    "    dataset_artifacts_folder=dataset_artifacts.download()\n",
    "    dataset = load_dataset('json', data_files=dataset_artifacts_folder+'/train_mpt_hhrlhf_49k_ja.json')\n",
    "    dataset = dataset[\"train\"]\n",
    "    dataset = dataset.select_columns(\"text\")\n",
    "    datasets.append(dataset)\n",
    "    # load OpenAssistant_oasst1\n",
    "    dataset_artifacts=run.use_artifact(run.config[\"tuning_data_artifacts\"][\"OpenAssistant_oasst1\"], type='dataset')\n",
    "    dataset_artifacts_folder=dataset_artifacts.download()\n",
    "    dataset = load_dataset('json', data_files=dataset_artifacts_folder+'/train_OpenAssistant_oasst1.json')\n",
    "    dataset = dataset[\"train\"]\n",
    "    dataset = dataset.select_columns(\"text\")\n",
    "    datasets.append(dataset)\n",
    "\n",
    "    # load databricks-dolly-15k-ja\n",
    "    dataset_artifacts=run.use_artifact(run.config[\"tuning_data_artifacts\"][\"databricks-dolly-15k-ja\"], type='dataset')\n",
    "    dataset_artifacts_folder=dataset_artifacts.download()\n",
    "    dataset = load_dataset('json', data_files=dataset_artifacts_folder+'/train_databricks-dolly-15k-ja.json')\n",
    "    dataset = dataset[\"train\"]\n",
    "    dataset = dataset.select_columns(\"text\")\n",
    "    datasets.append(dataset)\n",
    "    eot_token = tokenizer.eos_token\n",
    "\n",
    "    for i in range(len(datasets)):\n",
    "        datasets[i] = datasets[i].filter(lambda example: len(example[\"text\"]) <= 1000)\n",
    "        datasets[i] = datasets[i].map(lambda example: {\"text\": example[\"text\"] + eot_token})\n",
    "    \n",
    "    train_dataset = concatenate_datasets(datasets)\n",
    "    train_dataset = train_dataset.train_test_split(test_size=0.1)\n",
    "\n",
    "    #instruction_ids = tokenizer.encode(\"\\n\\n### 指示:\\n\", add_special_tokens=False)[1:]\n",
    "    #response_ids = tokenizer.encode(\"\\n\\n### 応答:\\n\", add_special_tokens=False)[1:]\n",
    "    instruction_template = \"### 指示:\"\n",
    "    response_template = \"### 応答:\"\n",
    "    collator = DataCollatorForCompletionOnlyLM(\n",
    "        instruction_template=instruction_template,response_template=response_template, tokenizer=tokenizer\n",
    "    )\n",
    "\n",
    "    trainer = SFTTrainer(\n",
    "        model,\n",
    "        args=transformers.TrainingArguments(**wandb.config[\"training_args\"]),\n",
    "        tokenizer=tokenizer,\n",
    "        train_dataset=train_dataset[\"train\"],\n",
    "        eval_dataset=train_dataset[\"test\"],\n",
    "        peft_config=lora_config,\n",
    "        dataset_text_field=\"text\",\n",
    "        max_seq_length=run.config[\"max_seq_length\"],\n",
    "        data_collator=collator,\n",
    "    )\n",
    "    trainer.train()\n",
    "    trainer.save_model()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "llm-finetuning",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.1.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}