{
  "_name_or_path": "./checkpoints/temporal-dattn_mistral-7b-v0.3_pt2-qwen_aud_high-res",
  "architectures": [
    "DattnMistralForCausalLM"
  ],
  "attention_dropout": 0.0,
  "bos_token_id": 1,
  "eos_token_id": 2,
  "head_dim": 128,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "initializer_range": 0.02,
  "intermediate_size": 14336,
  "loss_thres": 0.1,
  "max_position_embeddings": 32768,
  "mm_audio_pool_size": 5,
  "mm_audio_tower": "openai/whisper-large-v3",
  "mm_image_aspect_ratio": "resize",
  "mm_image_grid_points": null,
  "mm_image_pool_size": 5,
  "mm_input_type": "video",
  "mm_splits": 4,
  "mm_std": 0.0026716517750173807,
  "mm_time_interval": 10000,
  "mm_vision_select_layer": -2,
  "mm_vision_tower": "google/siglip-so400m-patch14-384",
  "model_type": "dattn_mistral",
  "num_attention_heads": 32,
  "num_hidden_layers": 32,
  "num_key_value_heads": 8,
  "pad_token_id": 0,
  "rms_norm_eps": 1e-05,
  "rope_theta": 1000000.0,
  "sliding_window": null,
  "tie_word_embeddings": false,
  "torch_dtype": "bfloat16",
  "train_aud": false,
  "train_llm": true,
  "train_rand": true,
  "train_vis": false,
  "transformers_version": "4.49.0",
  "use_cache": true,
  "vocab_size": 32768
}