| "architectures": [ | |
| "CognoSphereUnifiedMultimodalLanguageModel" | |
| ], | |
| "attention_probs_dropout_prob": 0.1, | |
| "classifier_dropout": null, # Optional; dropout probability for the classifier, if any | |
| "gradient_checkpointing": false, # Optional; enable gradient checkpointing to reduce memory usage | |
| "hidden_act": "gelu", | |
| "hidden_dropout_prob": 0.1, | |
| "hidden_size": 768, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 3072, | |
| "layer_norm_eps": 1e-12, | |
| "max_position_embeddings": 512, | |
| "model_type": "cogsphere", | |
| "num_attention_heads": 12, | |
| "num_hidden_layers": 12, | |
| "pad_token_id": 0, | |
| "position_embedding_type": "absolute", | |
| "transformers_version": "4.26.0", | |
| "type_vocab_size": 2, | |
| "vocab_size": 30522, | |
| "model_card": "https://huggingface.co/CognoSphere/CSUMLM/blob/main/README.md" | |
| } |