moe-minicpm-x4-base / config.json
babybirdprd's picture
Upload folder using huggingface_hub
6316cb1 verified
{
"_name_or_path": "NurtureAI/minicpm-2b-sft-bf16-llamafied-16k",
"architectures": [
"MixtralForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 1,
"dim_model_base": 256,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 2304,
"initializer_range": 0.1,
"intermediate_size": 5760,
"max_position_embeddings": 16384,
"model_type": "mixtral",
"num_attention_heads": 36,
"num_experts_per_tok": 2,
"num_hidden_layers": 40,
"num_key_value_heads": 36,
"num_local_experts": 4,
"output_router_logits": false,
"pretraining_tp": 1,
"rms_norm_eps": 1e-05,
"rope_scaling": {
"factor": 4.0,
"type": "linear"
},
"rope_theta": 10000.0,
"router_aux_loss_coef": 0.001,
"scale_depth": 1.4,
"scale_emb": 12,
"sliding_window": null,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.37.2",
"use_cache": true,
"vocab_size": 122753
}