MoE_4experts_topk2 / config.json
lukeysong's picture
Upload folder using huggingface_hub
c3ab4fc verified
{
"architectures": [
"LlamaForCausalLMEagle3MoE"
],
"attention_bias": false,
"attention_chunk_size": 8192,
"attention_dropout": 0.0,
"attn_scale": 0.1,
"attn_temperature_tuning": true,
"bos_token_id": 128000,
"draft_vocab_size": 32000,
"dtype": "bfloat16",
"eos_token_id": 128001,
"floor_scale": 8192,
"has_shared_expert": false,
"head_dim": 128,
"hidden_act": "silu",
"hidden_size": 4096,
"initializer_range": 0.02,
"interleave_moe_layer_step": 1,
"intermediate_size": 14336,
"intermediate_size_mlp": 16384,
"layer_types": [
"chunked_attention"
],
"max_position_embeddings": 2048,
"model_type": "llama4_text",
"moe_layers": [
0
],
"no_rope_layers": [
1
],
"num_attention_heads": 32,
"num_experts_per_tok": 2,
"num_hidden_layers": 1,
"num_key_value_heads": 8,
"num_local_experts": 4,
"output_router_logits": false,
"pad_token_id": 0,
"rms_norm_eps": 1e-05,
"rope_scaling": null,
"rope_theta": 500000,
"router_aux_loss_coef": 0.001,
"router_jitter_noise": 0.0,
"tie_word_embeddings": false,
"transformers_version": "4.57.1",
"use_cache": true,
"use_qk_norm": true,
"vocab_size": 128256
}