tinydeepseekv3-3m / deepseek_v3_config_dump.json
shibatch's picture
Upload folder using huggingface_hub
6036082 verified
Raw
History Blame Contribute Delete
1.47 kB
{
"transformers_version": "5.14.1",
"architectures": null,
"output_hidden_states": false,
"return_dict": true,
"dtype": null,
"chunk_size_feed_forward": 0,
"is_encoder_decoder": false,
"id2label": {
"0": "LABEL_0",
"1": "LABEL_1"
},
"label2id": {
"LABEL_0": 0,
"LABEL_1": 1
},
"problem_type": null,
"vocab_size": 1024,
"hidden_size": 216,
"intermediate_size": 432,
"moe_intermediate_size": 128,
"num_hidden_layers": 5,
"num_attention_heads": 8,
"num_key_value_heads": 8,
"n_shared_experts": 1,
"n_routed_experts": 4,
"routed_scaling_factor": 2.5,
"kv_lora_rank": 64,
"q_lora_rank": 64,
"qk_rope_head_dim": 16,
"v_head_dim": 32,
"qk_nope_head_dim": 16,
"n_group": 1,
"topk_group": 1,
"num_experts_per_tok": 1,
"first_k_dense_replace": 1,
"norm_topk_prob": true,
"hidden_act": "silu",
"max_position_embeddings": 2048,
"initializer_range": 0.02,
"rms_norm_eps": 1e-06,
"use_cache": false,
"pad_token_id": 1000,
"bos_token_id": 1000,
"eos_token_id": 1001,
"pretraining_tp": 1,
"tie_word_embeddings": true,
"rope_parameters": {
"rope_type": "default",
"rope_theta": 10000
},
"rope_interleave": true,
"attention_bias": false,
"attention_dropout": 0.0,
"num_nextn_predict_layers": 0,
"qk_head_dim": 32,
"head_dim": 16,
"_name_or_path": "",
"aux_loss_alpha": 0.0001,
"seq_aux": true,
"model_type": "deepseek_v3",
"output_attentions": false
}