Why model 8B is shown as 7B ?

[Piotr-rogal]

Describe the bug

I use LM Studio v.0.2.31 and when i download:
https://huggingface.co/lmstudio-community/Meta-Llama-3.1-8B-Instruct-GGUF/resolve/main/Meta-Llama-3.1-8B-Instruct-Q4_K_M.gguf
or
meta-llama/Meta-Llama-3.1-8B-Instruct
it shows model 7B not 8B.

Is this a bug in LM Studio or in model ?

If i look in model inspector they show:

{
"name": "Meta Llama 3.1 8B Instruct",
"arch": "llama",
"quant": "Q4_K_M",
"context_length": 131072,
"embedding_length": 4096,
"num_layers": 32,
"rope": {
"freq_base": 500000,
"dimension_count": 128
},
"head_count": 32,
"head_count_kv": 8,
"parameters": "7B"
}

I try to download many 8B models (marked as 8B) and always finaly are 7B

Have any idea why ?