vllm_hairuo/ihp/config/model/cfg.hairuo.2b.json
2024-10-25 17:16:26 +08:00

38 lines
1015 B
JSON

{
"config_clz": "hairuo.HairuoConfig",
"config": {
"attention_dropout": 0.0,
"bos_token_id": 151643,
"eos_token_id": 151645,
"hidden_act": "silu",
"hidden_size": 2304,
"initializer_range": 0.02,
"intermediate_size": 5760,
"max_position_embeddings": 131072,
"model_type": "hairuo",
"num_attention_heads": 36,
"num_hidden_layers": 40,
"num_key_value_heads": 36,
"rms_norm_eps": 1e-05,
"rope_scaling": {
"factor": 8.0,
"high_freq_factor": 4.0,
"low_freq_factor": 1.0,
"original_max_position_embeddings": 8192,
"rope_type": "llama3"
},
"rope_theta": 500000.0,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"use_cache": true,
"vocab_size": 152064,
"_attn_implementation": "eager",
"_flash_attn_2_enabled": false,
"mup_scale_emb": 12,
"mup_scale_depth": 1.4,
"mup_scale_width": 9.0
},
"model_clz": "hairuo.HairuoForCausalLM",
"tokenizer_clz": "hairuo.HairuoTokenizer"
}