GK-MoE-0.1 / config.json
powermove72's picture
Upload folder using huggingface_hub
a08b0e9 verified
{
"_name_or_path": "GritLM/GritLM-7B",
"architectures": [
"MixtralForCausalLM"
],
"attention_dropout": 0.0,
"auto_map": {
"AutoModel": "GritLM/GritLM-7B--modeling_gritlm7b.MistralModel",
"AutoModelForCausalLM": "GritLM/GritLM-7B--modeling_gritlm7b.MistralForCausalLM",
"AutoModelForSequenceClassification": "GritLM/GritLM-7B--modeling_gritlm7b.MistralForSequenceClassification"
},
"bos_token_id": 1,
"eos_token_id": 2,
"hidden_act": "silu",
"hidden_size": 4096,
"id2label": {
"0": "LABEL_0"
},
"initializer_range": 0.02,
"intermediate_size": 14336,
"label2id": {
"LABEL_0": 0
},
"max_position_embeddings": 32768,
"model_type": "mixtral",
"num_attention_heads": 32,
"num_experts_per_tok": 2,
"num_hidden_layers": 32,
"num_key_value_heads": 8,
"num_local_experts": 2,
"output_router_logits": false,
"rms_norm_eps": 1e-05,
"rope_theta": 10000.0,
"router_aux_loss_coef": 0.001,
"router_jitter_noise": 0.0,
"sliding_window": null,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.41.2",
"use_cache": true,
"vocab_size": 32000
}