codewithdark/DiffusionLM
136
1{2 "architectures": [3 "DiffusionLLM"4 ],5 "auto_map": {6 "AutoModel": "modeling_DiffusionLLM.DiffusionLLM",7 "AutoModelForCausalLM": "modeling_DiffusionLLM.DiffusionLLM",8 "AutoConfig": "modeling_DiffusionLLM.DiffusionConfig"9 },10 "attention_probs_dropout_prob": 0.1,11 "eos_token_id": 50256,12 "hidden_dropout_prob": 0.1,13 "hidden_size": 256,14 "initializer_range": 0.02,15 "intermediate_size": 3072,16 "layer_norm_eps": 1e-12,17 "mask_token_id": 50257,18 "max_position_embeddings": 256,19 "model_type": "diffusionLM",20 "num_attention_heads": 4,21 "num_hidden_layers": 4,22 "num_timesteps": 100,23 "pad_token_id": 50258,24 "time_embed_dim": 128,25 "torch_dtype": "float32",26 "transformers_version": "4.48.3",27 "vocab_size": 5025928}29 