Nova-13B-50-step / config.json
Q-bert's picture
Upload folder using huggingface_hub (#1)
1a827cc
raw
history blame
742 Bytes
{
"_name_or_path": "/root/.cache/huggingface/hub/models--AIDC-ai-business--Luban-13B/snapshots/72d10ef532610a5bae01c24d7a0964949acc2a57",
"architectures": [
"LlamaForCausalLM"
],
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 5120,
"initializer_range": 0.02,
"intermediate_size": 13824,
"max_position_embeddings": 4096,
"model_type": "llama",
"num_attention_heads": 40,
"num_hidden_layers": 40,
"num_key_value_heads": 40,
"pad_token_id": 0,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 10000.0,
"tie_word_embeddings": false,
"torch_dtype": "float16",
"transformers_version": "4.33.0",
"use_cache": false,
"vocab_size": 32002
}