MiniCPM-MoE-8x2B / config.json
finalf0's picture
add model files
7d01e83
raw
history blame contribute delete
998 Bytes
{
"_name_or_path": "openbmb/CPM-2B",
"architectures": [
"MiniCPMForCausalLM"
],
"auto_map": {
"AutoConfig": "configuration_minicpm.MiniCPMConfig",
"AutoModel": "modeling_minicpm.MiniCPMModel",
"AutoModelForCausalLM": "modeling_minicpm.MiniCPMForCausalLM",
"AutoModelForSeq2SeqLM": "modeling_minicpm.MiniCPMForCausalLM",
"AutoModelForSequenceClassification": "modeling_minicpm.MiniCPMForSequenceClassification"
},
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 2304,
"initializer_range": 0.1,
"intermediate_size": 5760,
"max_position_embeddings": 4096,
"num_attention_heads": 36,
"num_hidden_layers": 40,
"num_key_value_heads": 36,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"torch_dtype": "bfloat16",
"transformers_version": "4.36.0",
"use_cache": true,
"vocab_size": 122753,
"scale_emb": 12,
"dim_model_base": 256,
"scale_depth": 1.4,
"num_experts": 8,
"num_experts_per_tok": 2
}