31 lines
873 B
JSON
31 lines
873 B
JSON
{
|
|
"_name_or_path": "/scratch3/he051/u_dpo/runs/sft/hh-rlhf/Qwen2.5-7B_right-padding/adamw/lr-1e-05_batch_size-128/epoch-1/weight_decay-0/seed-42/checkpoint-1197/",
|
|
"architectures": [
|
|
"Qwen2ForCausalLM"
|
|
],
|
|
"attention_dropout": 0.0,
|
|
"bos_token_id": 151645,
|
|
"eos_token_id": 151645,
|
|
"hidden_act": "silu",
|
|
"hidden_size": 3584,
|
|
"initializer_range": 0.02,
|
|
"intermediate_size": 18944,
|
|
"max_position_embeddings": 131072,
|
|
"max_window_layers": 28,
|
|
"model_type": "qwen2",
|
|
"num_attention_heads": 28,
|
|
"num_hidden_layers": 28,
|
|
"num_key_value_heads": 4,
|
|
"rms_norm_eps": 1e-06,
|
|
"rope_scaling": null,
|
|
"rope_theta": 1000000.0,
|
|
"sliding_window": null,
|
|
"tie_word_embeddings": false,
|
|
"torch_dtype": "bfloat16",
|
|
"transformers_version": "4.47.1",
|
|
"use_cache": false,
|
|
"use_mrope": false,
|
|
"use_sliding_window": false,
|
|
"vocab_size": 152064
|
|
}
|