whisp / config.json
thewh1teagle's picture
Upload folder using huggingface_hub
4897b34 verified
{
"architectures": [
"Qwen3MoeForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 2,
"decoder_sparse_step": 1,
"dtype": "float32",
"eos_token_id": 3,
"hidden_act": "silu",
"hidden_size": 384,
"initializer_range": 0.02,
"intermediate_size": 1024,
"max_position_embeddings": 4096,
"mlp_only_layers": [
0,
1,
2,
3,
4,
5
],
"model_type": "qwen3_moe",
"moe_intermediate_size": 1024,
"norm_topk_prob": false,
"num_attention_heads": 6,
"num_experts_per_tok": 1,
"num_hidden_layers": 12,
"num_key_value_heads": 6,
"num_local_experts": 32,
"output_router_logits": false,
"pad_token_id": 0,
"rms_norm_eps": 1e-06,
"rope_parameters": {
"rope_theta": 500000.0,
"rope_type": "default"
},
"router_aux_loss_coef": 0.001,
"sliding_window": null,
"tie_word_embeddings": true,
"transformers_version": "5.7.0",
"use_cache": true,
"use_sliding_window": false,
"vocab_size": 10455
}