dots-studio/dots.llm1.base
699.5k
1{2 "architectures": [3 "Dots1ForCausalLM"4 ],5 "attention_bias": false,6 "attention_dropout": 0.0,7 "bos_token_id": null,8 "eos_token_id": 151643,9 "first_k_dense_replace": 1,10 "hidden_act": "silu",11 "hidden_size": 4096,12 "initializer_range": 0.02,13 "intermediate_size": 10944,14 "max_position_embeddings": 32768,15 "model_type": "dots1",16 "moe_intermediate_size": 1408,17 "moe_layer_freq": 1,18 "n_routed_experts": 128,19 "n_shared_experts": 2,20 "norm_topk_prob": true,21 "num_attention_heads": 32,22 "num_experts_per_tok": 6,23 "num_hidden_layers": 62,24 "num_key_value_heads": 32,25 "pretraining_tp": 1,26 "rms_norm_eps": 1e-05,27 "rope_scaling": null,28 "rope_theta": 10000000,29 "routed_scaling_factor": 2.5,30 "scoring_func": "sigmoid",31 "sliding_window": null,32 "topk_method": "noaux_tc",33 "tie_word_embeddings": false,34 "torch_dtype": "bfloat16",35 "transformers_version": "4.46.3",36 "use_cache": true,37 "use_sliding_window": false,38 "vocab_size": 15206439}40 