tttx/models-ttt-diff-buffer-step1-021325
03
1{2 "_attn_implementation_autoset": true,3 "_name_or_path": "deepseek-ai/DeepSeek-R1-Distill-Qwen-32B",4 "architectures": [5 "Qwen2ForCausalLM"6 ],7 "attention_dropout": 0.0,8 "bos_token_id": 151643,9 "eos_token_id": 151643,10 "hidden_act": "silu",11 "hidden_size": 5120,12 "initializer_range": 0.02,13 "intermediate_size": 27648,14 "max_position_embeddings": 131072,15 "max_window_layers": 64,16 "model_type": "qwen2",17 "num_attention_heads": 40,18 "num_hidden_layers": 64,19 "num_key_value_heads": 8,20 "rms_norm_eps": 1e-05,21 "rope_scaling": null,22 "rope_theta": 1000000.0,23 "sliding_window": null,24 "tie_word_embeddings": false,25 "torch_dtype": "bfloat16",26 "transformers_version": "4.47.0.dev0",27 "use_cache": true,28 "use_sliding_window": false,29 "vocab_size": 15206430}31 