princeton-nlp's picture
Update config.json
43c5122
raw
history blame
750 Bytes
{
"_name_or_path": "/scratch/gpfs/mengzhou/space2/out/test_round23_mosaicml_version5/llama2_7b_pruning_scaling_doremi_to1.3b_sl4096_finetune_bs1M_fullrun/checkpoint-51200",
"architectures": [
"LlamaForCausalLM"
],
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 2048,
"initializer_range": 0.02,
"intermediate_size": 5504,
"max_position_embeddings": 4096,
"model_type": "llama",
"num_attention_heads": 16,
"num_hidden_layers": 24,
"num_key_value_heads": 16,
"pad_token_id": 0,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"tie_word_embeddings": false,
"torch_dtype": "float32",
"transformers_version": "4.28.0",
"use_cache": true,
"vocab_size": 32001
}