DLight1551's picture
add new model
24df943
raw
history blame contribute delete
954 Bytes
{
"architectures": [
"InternLM2ForCausalLM"
],
"attn_implementation": "flash_attention_2",
"auto_map": {
"AutoConfig": "configuration_internlm_xcomposer2.InternLMXcomposer2Config",
"AutoModel": "modeling_internlm_xcomposer2.InternLMXComposer2ForCausalLM",
"AutoModelForCausalLM": "modeling_internlm_xcomposer2.InternLMXComposer2ForCausalLM"
},
"bias": false,
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 4096,
"initializer_range": 0.02,
"intermediate_size": 14336,
"max_length": 16384,
"max_position_embeddings": 32768,
"model_type": "internlmxcomposer2",
"num_attention_heads": 32,
"num_hidden_layers": 32,
"num_key_value_heads": 8,
"pad_token_id": 2,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 1000000,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.33.1",
"use_cache": false,
"vocab_size": 92544
}