jungyuko's picture
Upload LlamaForCausalLM
c1f0d70 verified
raw
history blame
875 Bytes
{
"_name_or_path": "/home/aift-ml3/workspace2/GAI/nlp/llm_research/train/KoAlpaca/train_v1.1b/AIFT-42dot_LLM-PLM-1.3B-ao-instruct-all-v1.1.2-ff-e3/checkpoint-379",
"architectures": [
"LlamaForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 50257,
"eos_token_id": 50256,
"hidden_act": "silu",
"hidden_size": 2048,
"initializer_range": 0.01,
"intermediate_size": 5632,
"max_position_embeddings": 4096,
"model_type": "llama",
"num_attention_heads": 32,
"num_hidden_layers": 24,
"num_key_value_heads": 32,
"pad_token_id": 50258,
"pretraining_tp": 1,
"rms_norm_eps": 1e-06,
"rope_scaling": null,
"rope_theta": 10000.0,
"tie_word_embeddings": false,
"tokenizer_class": "GPT2TokenizerFast",
"torch_dtype": "float32",
"transformers_version": "4.36.2",
"use_cache": true,
"vocab_size": 50304
}