hmellor/tiny-random-DeepseekV2ForCausalLM
021k
1{2 "architectures": [3 "DeepseekV2ForCausalLM"4 ],5 "attention_bias": false,6 "attention_dropout": 0.0,7 "bos_token_id": 100000,8 "dtype": "bfloat16",9 "eos_token_id": 100001,10 "first_k_dense_replace": 1,11 "head_dim": 64,12 "hidden_act": "silu",13 "hidden_size": 256,14 "initializer_range": 0.02,15 "intermediate_size": 512,16 "kv_lora_rank": 512,17 "max_position_embeddings": 163840,18 "mlp_bias": false,19 "model_type": "deepseek_v2",20 "moe_intermediate_size": 128,21 "moe_layer_freq": 1,22 "n_group": 1,23 "n_routed_experts": 8,24 "n_shared_experts": 2,25 "norm_topk_prob": false,26 "num_attention_heads": 4,27 "num_experts_per_tok": 2,28 "num_hidden_layers": 4,29 "num_key_value_heads": 4,30 "pad_token_id": null,31 "pretraining_tp": 1,32 "q_lora_rank": null,33 "qk_head_dim": 192,34 "qk_nope_head_dim": 128,35 "qk_rope_head_dim": 64,36 "rms_norm_eps": 1e-06,37 "rope_parameters": {38 "beta_fast": 32,39 "beta_slow": 1,40 "factor": 40,41 "mscale": 0.707,42 "mscale_all_dim": 0.707,43 "original_max_position_embeddings": 4096,44 "rope_theta": 10000,45 "rope_type": "yarn",46 "type": "yarn"47 },48 "routed_scaling_factor": 1.0,49 "scoring_func": "softmax",50 "tie_word_embeddings": false,51 "topk_group": 1,52 "topk_method": "greedy",53 "transformers_version": "5.15.0.dev0",54 "use_cache": true,55 "v_head_dim": 128,56 "vocab_size": 10240057}58 