deepseek-moe-16b-base / config.json
vltnmmdv's picture
Upload DeepseekFixedForCausalLM
4bb2538 verified
{
"_name_or_path": "deepseek-ai/deepseek-moe-16b-base",
"architectures": [
"DeepseekFixedForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"auto_map": {
"AutoConfig": "configuration_deepseek_fixed.DeepseekFixedConfig",
"AutoModel": "deepseek-ai/deepseek-moe-16b-base--modeling_deepseek.DeepseekModel",
"AutoModelForCausalLM": "modelling_deepseek_fixed.DeepseekFixedForCausalLM"
},
"aux_loss_alpha": 0.001,
"bos_token_id": 100000,
"eos_token_id": 100001,
"first_k_dense_replace": 1,
"hidden_act": "silu",
"hidden_size": 2048,
"initializer_range": 0.02,
"intermediate_size": 10944,
"max_position_embeddings": 4096,
"model_type": "deepseek_with_concentration",
"moe_implementation": "eager",
"moe_intermediate_size": 1408,
"moe_layer_freq": 1,
"n_routed_experts": 64,
"n_shared_experts": 2,
"norm_topk_prob": false,
"num_attention_heads": 16,
"num_experts_per_tok": 6,
"num_hidden_layers": 28,
"num_key_value_heads": 16,
"pretraining_tp": 1,
"rms_norm_eps": 1e-06,
"rope_scaling": null,
"rope_theta": 10000,
"scoring_func": "softmax",
"seq_aux": true,
"tie_word_embeddings": false,
"torch_dtype": "float32",
"transformers_version": "4.43.3",
"use_cache": true,
"vocab_size": 102400
}