KwaiCoder-23B-A4B-v1 / config.json
zhangxiaojiang
bf16 commit
18030b7
raw
history blame contribute delete
795 Bytes
{
"activation": "silu",
"architectures": [
"MixtralForCausalLM"
],
"attention_dropout": 0.0,
"bos_token_id": 32013,
"eos_token_id": 32014,
"head_dim": 128,
"hidden_act": "silu",
"hidden_size": 4096,
"initializer_range": 0.02,
"intermediate_size": 1376,
"max_position_embeddings": 4096,
"model_type": "mixtral",
"num_attention_heads": 32,
"num_experts_per_tok": 8,
"num_hidden_layers": 20,
"num_key_value_heads": 32,
"num_local_experts": 64,
"output_router_logits": false,
"rms_norm_eps": 1e-06,
"rope_theta": 100000.0,
"router_aux_loss_coef": 0.02,
"router_jitter_noise": 0.0,
"sliding_window": null,
"tie_word_embeddings": false,
"torch_dtype": "bfloat16",
"transformers_version": "4.48.1",
"use_cache": true,
"vocab_size": 32128
}