Ziya-Coding-15B-v1 / config.json
pskun's picture
Upload folder using huggingface_hub (#1)
78649a6
raw
history blame
698 Bytes
{
"_name_or_path": "/cognitive_comp/pankunhao/pretrained/Ziya-Coding-v1",
"activation_function": "gelu",
"architectures": [
"GPTBigCodeForCausalLM"
],
"attention_softmax_in_fp32": true,
"attn_pdrop": 0.1,
"bos_token_id": 0,
"embd_pdrop": 0.1,
"eos_token_id": 0,
"initializer_range": 0.02,
"layer_norm_epsilon": 1e-05,
"model_type": "gpt_bigcode",
"multi_query": true,
"n_embd": 6144,
"n_head": 48,
"n_inner": 24576,
"n_layer": 40,
"n_positions": 8192,
"resid_pdrop": 0.1,
"scale_attention_softmax_in_fp32": true,
"scale_attn_weights": true,
"torch_dtype": "bfloat16",
"transformers_version": "4.32.0",
"use_cache": false,
"vocab_size": 49155
}