| { |
| "architectures": [ |
| "LlavaForConditionalGeneration" |
| ], |
| "ignore_index": -100, |
| "image_token_index": 32022, |
| "model_type": "llava", |
| "projector_hidden_act": "gelu", |
| "text_config": { |
| "_name_or_path": "LLM360/K2-Chat", |
| "architectures": [ |
| "LlamaForCausalLM" |
| ], |
| "bos_token_id": null, |
| "hidden_size": 8192, |
| "intermediate_size": 22016, |
| "max_position_embeddings": 8192, |
| "max_sequence_length": 8192, |
| "model_type": "llama", |
| "num_attention_heads": 64, |
| "num_hidden_layers": 80, |
| "num_key_value_heads": 64, |
| "pad_token_id": 0, |
| "rms_norm_eps": 1e-05, |
| "rope_theta": 500000.0, |
| "torch_dtype": "float16", |
| "vocab_size": 32032 |
| }, |
| "torch_dtype": "float32", |
| "transformers_version": "4.42.4", |
| "vision_config": { |
| "architectures": [ |
| "CLIPVisionModel" |
| ], |
| "dropout": 0.0, |
| "hidden_size": 1024, |
| "image_size": 336, |
| "intermediate_size": 4096, |
| "model_type": "clip_vision_model", |
| "num_attention_heads": 16, |
| "num_hidden_layers": 24, |
| "patch_size": 14, |
| "projection_dim": 768 |
| }, |
| "vision_feature_layer": -2, |
| "vision_feature_select_strategy": "default" |
| } |
|
|
