Safetensors
llama
zqh11's picture
upload models
c7ea101
raw
history blame
1.38 kB
{
"architectures": [
"DeepseekForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"auto_map": {
"AutoConfig": "configuration_deepseek.DeepseekConfig",
"AutoModel": "modeling_deepseek.DeepseekModel",
"AutoModelForCausalLM": "modeling_deepseek.DeepseekForCausalLM"
},
"aux_loss_alpha": 0.001,
"bos_token_id": 100000,
"eos_token_id": 100001,
"ep_size": 1,
"first_k_dense_replace": null,
"hidden_act": "silu",
"hidden_size": 4096,
"initializer_range": 0.02,
"intermediate_size": 11008,
"kv_lora_rank": null,
"max_position_embeddings": 4096,
"model_type": "deepseek",
"moe_intermediate_size": 11008,
"moe_layer_freq": null,
"n_group": 1,
"n_routed_experts": null,
"n_shared_experts": null,
"norm_topk_prob": null,
"num_attention_heads": 32,
"num_experts_per_tok": null,
"num_hidden_layers": 30,
"num_key_value_heads": 32,
"pretraining_tp": 1,
"q_lora_rank": null,
"qk_nope_head_dim": 0,
"qk_rope_head_dim": null,
"rms_norm_eps": 1e-06,
"rope_scaling": null,
"rope_theta": 10000,
"routed_scaling_factor": 1.0,
"scoring_func": null,
"seq_aux": true,
"tie_word_embeddings": false,
"topk_group": 1,
"topk_method": "group_limited_greedy",
"torch_dtype": "bfloat16",
"transformers_version": "4.33.1",
"use_cache": true,
"v_head_dim": null,
"vocab_size": 102400
}