| { | |
| "add_faster_video": false, | |
| "add_time_instruction": false, | |
| "architectures": [ | |
| "LlavaLfm2ForCausalLM" | |
| ], | |
| "block_auto_adjust_ff_dim": true, | |
| "block_dim": 1024, | |
| "block_ff_dim": 6656, | |
| "block_ffn_dim_multiplier": 1.0, | |
| "block_mlp_init_scale": 1.0, | |
| "block_multiple_of": 256, | |
| "block_norm_eps": 1e-05, | |
| "block_out_init_scale": 1.0, | |
| "block_use_swiglu": true, | |
| "block_use_xavier_init": true, | |
| "bos_token_id": 1, | |
| "conv_L_cache": 3, | |
| "conv_bias": false, | |
| "conv_dim": 1024, | |
| "conv_dim_out": 1024, | |
| "conv_use_xavier_init": true, | |
| "dtype": "bfloat16", | |
| "eos_token_id": 7, | |
| "faster_token_stride": 10, | |
| "force_sample": false, | |
| "hidden_size": 1024, | |
| "image_aspect_ratio": "convnext", | |
| "image_crop_resolution": null, | |
| "image_grid_pinpoints": null, | |
| "image_split_resolution": null, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 6656, | |
| "layer_types": [ | |
| "conv", | |
| "conv", | |
| "full_attention", | |
| "conv", | |
| "conv", | |
| "full_attention", | |
| "conv", | |
| "conv", | |
| "full_attention", | |
| "conv", | |
| "full_attention", | |
| "conv", | |
| "full_attention", | |
| "conv", | |
| "full_attention", | |
| "conv" | |
| ], | |
| "max_position_embeddings": 128000, | |
| "mm_hidden_size": 2048, | |
| "mm_newline_position": "grid", | |
| "mm_patch_merge_type": "flat", | |
| "mm_projector_lr": 5e-07, | |
| "mm_projector_type": "layernorm", | |
| "mm_resampler_type": "identity", | |
| "mm_spatial_pool_mode": "bilinear", | |
| "mm_spatial_pool_out_channels": null, | |
| "mm_spatial_pool_stride": null, | |
| "mm_tunable_parts": "mm_mlp_adapter, mm_vision_tower, mm_language_model", | |
| "mm_use_im_patch_token": false, | |
| "mm_use_im_start_end": false, | |
| "mm_vision_resolution": 1536, | |
| "mm_vision_select_feature": "patch", | |
| "mm_vision_select_layer": -1, | |
| "mm_vision_tower": "/lustre/scratch/client/movian/research/users/chitb/ELVIS_SAVE_MODELS/converted_convnext_base_w_320", | |
| "mm_vision_tower_lr": 5e-07, | |
| "model_type": "lfm2", | |
| "norm_eps": 1e-05, | |
| "num_attention_heads": 16, | |
| "num_heads": 16, | |
| "num_hidden_layers": 16, | |
| "num_key_value_heads": 8, | |
| "pad_token_id": 0, | |
| "pos_skipping_range": 4096, | |
| "rope_scaling": null, | |
| "rope_theta": 1000000.0, | |
| "tokenizer_model_max_length": 8192, | |
| "tokenizer_padding_side": "right", | |
| "transformers_version": "4.57.1", | |
| "use_cache": false, | |
| "use_mm_proj": true, | |
| "use_pos_enc": true, | |
| "use_pos_skipping": false, | |
| "vision_add_five_stage": 1, | |
| "vision_five_stage_width": 2048, | |
| "vision_tower_pretrained": null, | |
| "vocab_size": 65536 | |
| } | |