tiny-random/gpt-oss-bf16
138k
1{2 "architectures": [3 "GptOssForCausalLM"4 ],5 "attention_bias": true,6 "attention_dropout": 0.0,7 "eos_token_id": 200002,8 "experts_per_token": 4,9 "head_dim": 32,10 "hidden_act": "silu",11 "hidden_size": 32,12 "initial_context_length": 4096,13 "initializer_range": 0.02,14 "intermediate_size": 64,15 "layer_types": [16 "sliding_attention",17 "full_attention"18 ],19 "max_position_embeddings": 131072,20 "model_type": "gpt_oss",21 "num_attention_heads": 2,22 "num_experts_per_tok": 4,23 "num_hidden_layers": 2,24 "num_key_value_heads": 1,25 "num_local_experts": 32,26 "output_router_logits": false,27 "pad_token_id": 199999,28 "rms_norm_eps": 1e-05,29 "rope_scaling": {30 "beta_fast": 32.0,31 "beta_slow": 1.0,32 "factor": 32.0,33 "original_max_position_embeddings": 4096,34 "rope_type": "yarn",35 "truncate": false36 },37 "rope_theta": 150000,38 "router_aux_loss_coef": 0.9,39 "sliding_window": 128,40 "swiglu_limit": 7.0,41 "tie_word_embeddings": true,42 "torch_dtype": "bfloat16",43 "transformers_version": "4.56.0.dev0",44 "use_cache": true,45 "vocab_size": 20108846}47 