| { | |
| "_name_or_path": "mistralai/Mistral-7B-v0.3", | |
| "architectures": [ | |
| "MistralForDiffusionLM" | |
| ], | |
| "attention_dropout": 0.0, | |
| "bos_token_id": 1, | |
| "classifier_free_simplex_inputs": false, | |
| "classifier_free_uncond_input": "empty_token", | |
| "deepmind_conditional": false, | |
| "disable_timestep_embed": false, | |
| "empty_token_be_mask": false, | |
| "eos_token_id": 2, | |
| "hidden_act": "silu", | |
| "hidden_size": 4096, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 14336, | |
| "is_causal": false, | |
| "mask_padding_in_loss": false, | |
| "max_position_embeddings": 32768, | |
| "model_type": "mistral", | |
| "num_attention_heads": 32, | |
| "num_hidden_layers": 32, | |
| "num_key_value_heads": 8, | |
| "pad_token_id": 32768, | |
| "padding_side": "right", | |
| "rms_norm_eps": 1e-05, | |
| "rope_theta": 1000000.0, | |
| "self_condition": "logits_mean", | |
| "self_condition_mix_before_weights": true, | |
| "self_condition_mix_logits_before_weights": false, | |
| "self_condition_mlp_projection": false, | |
| "self_condition_zeros_after_softmax": false, | |
| "sliding_window": null, | |
| "tie_word_embeddings": false, | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": "4.37.0", | |
| "use_cache": true, | |
| "vocab_size": 32769 | |
| } | |