| { | |
| "architectures": [ | |
| "YuanVLChatModel" | |
| ], | |
| "attention_dropout": 0.0, | |
| "auto_map": { | |
| "AutoConfig": "configuration_yuanvl.YuanVLChatConfig", | |
| "AutoModel": "modeling_yuanvl_chat.YuanVLChatModel", | |
| "AutoModelForCausalLM": "modeling_yuanvl_chat.YuanVLChatModel" | |
| }, | |
| "clip_download_path": "path/to/internvit-448", | |
| "clip_model_name": "InternViT-448", | |
| "downsample_ratio": 0.5, | |
| "dynamic_image_size": true, | |
| "eos_token_id": 77185, | |
| "force_image_size": 448, | |
| "imagemlp_recompute": true, | |
| "img_context_token_id": 77188, | |
| "llm_config": { | |
| "_from_model_config": true, | |
| "architectures": [ | |
| "YuanForCausalLM" | |
| ], | |
| "attention_projection_size": 4096, | |
| "attn_dropout": 0.0, | |
| "attn_mask_type": "causal", | |
| "auto_map": { | |
| "AutoConfig": "configuration_yuanvl.YuanConfig", | |
| "AutoModelForCausalLM": "yuanvl.YuanForCausalLM" | |
| }, | |
| "bos_token": "<BOS>", | |
| "bos_token_id": 134960, | |
| "causal_mask": true, | |
| "dropout": 0, | |
| "eod_token": "<eod>", | |
| "eod_token_id": 77185, | |
| "ffn_hidden_size": 8192, | |
| "head_dim": 256, | |
| "hidden_act": "silu", | |
| "hidden_size": 2048, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 8192, | |
| "lf_conv2d_group": 1, | |
| "lf_conv2d_num_pad": 0, | |
| "mask_token_id": 77185, | |
| "max_position_embeddings": 131072, | |
| "model_max_length": 131072, | |
| "model_type": "yuan2", | |
| "moe_config": { | |
| "ffn_hidden_size": 8192, | |
| "gated_linear_unit": true, | |
| "moe_num_experts": 32, | |
| "moe_top_k": 2, | |
| "norm_topk_prob": true, | |
| "num_attention_router_heads": 1, | |
| "router_type": "attn_router" | |
| }, | |
| "num_attention_heads": 16, | |
| "num_hidden_layers": 24, | |
| "output_router_logits": true, | |
| "pad_token_id": 77188, | |
| "rep_token": "<rep>", | |
| "rep_token_id": 134962, | |
| "reset_attention_mask": false, | |
| "reset_position_ids": false, | |
| "rms_norm_eps": 1e-06, | |
| "rotary_base": 1000000, | |
| "rotary_percent": 0.5, | |
| "sep_token": "<sep>", | |
| "sep_token_id": 77187, | |
| "tie_word_embeddings": false, | |
| "tokenizer_class": "YuanVLTokenizer", | |
| "torch_dtype": "bfloat16", | |
| "use_bias": false, | |
| "use_cache": true, | |
| "use_flash_attention": true, | |
| "use_lf_gate": true, | |
| "use_lfa_bias": true, | |
| "use_loss_mask": false, | |
| "use_moe": true, | |
| "use_rope_scaling": false, | |
| "vocab_size": 136064 | |
| }, | |
| "max_dynamic_patch": 9, | |
| "min_dynamic_patch": 1, | |
| "model_type": "yuanvl", | |
| "output_attentions": false, | |
| "ps_version": "v2", | |
| "quantization_config": { | |
| "config_groups": { | |
| "group_0": { | |
| "format": "pack-quantized", | |
| "input_activations": null, | |
| "output_activations": null, | |
| "targets": [ | |
| "Linear" | |
| ], | |
| "weights": { | |
| "actorder": null, | |
| "block_structure": null, | |
| "dynamic": false, | |
| "group_size": 128, | |
| "num_bits": 4, | |
| "observer": "minmax", | |
| "observer_kwargs": {}, | |
| "strategy": "group", | |
| "symmetric": true, | |
| "type": "int" | |
| } | |
| } | |
| }, | |
| "format": "pack-quantized", | |
| "global_compression_ratio": null, | |
| "ignore": [ | |
| ".encoder.layers.0.attn.qkv", | |
| ".encoder.layers.0.attn.proj", | |
| ".encoder.layers.0.mlp.fc1", | |
| ".encoder.layers.0.mlp.fc2", | |
| ".encoder.layers.1.attn.qkv", | |
| ".encoder.layers.1.attn.proj", | |
| ".encoder.layers.1.mlp.fc1", | |
| ".encoder.layers.1.mlp.fc2", | |
| ".encoder.layers.2.attn.qkv", | |
| ".encoder.layers.2.attn.proj", | |
| ".encoder.layers.2.mlp.fc1", | |
| ".encoder.layers.2.mlp.fc2", | |
| ".encoder.layers.3.attn.qkv", | |
| ".encoder.layers.3.attn.proj", | |
| ".encoder.layers.3.mlp.fc1", | |
| ".encoder.layers.3.mlp.fc2", | |
| ".encoder.layers.4.attn.qkv", | |
| ".encoder.layers.4.attn.proj", | |
| ".encoder.layers.4.mlp.fc1", | |
| ".encoder.layers.4.mlp.fc2", | |
| ".encoder.layers.5.attn.qkv", | |
| ".encoder.layers.5.attn.proj", | |
| ".encoder.layers.5.mlp.fc1", | |
| ".encoder.layers.5.mlp.fc2", | |
| ".encoder.layers.6.attn.qkv", | |
| ".encoder.layers.6.attn.proj", | |
| ".encoder.layers.6.mlp.fc1", | |
| ".encoder.layers.6.mlp.fc2", | |
| ".encoder.layers.7.attn.qkv", | |
| ".encoder.layers.7.attn.proj", | |
| ".encoder.layers.7.mlp.fc1", | |
| ".encoder.layers.7.mlp.fc2", | |
| ".encoder.layers.8.attn.qkv", | |
| ".encoder.layers.8.attn.proj", | |
| ".encoder.layers.8.mlp.fc1", | |
| ".encoder.layers.8.mlp.fc2", | |
| ".encoder.layers.9.attn.qkv", | |
| ".encoder.layers.9.attn.proj", | |
| ".encoder.layers.9.mlp.fc1", | |
| ".encoder.layers.9.mlp.fc2", | |
| ".encoder.layers.10.attn.qkv", | |
| ".encoder.layers.10.attn.proj", | |
| ".encoder.layers.10.mlp.fc1", | |
| ".encoder.layers.10.mlp.fc2", | |
| ".encoder.layers.11.attn.qkv", | |
| ".encoder.layers.11.attn.proj", | |
| ".encoder.layers.11.mlp.fc1", | |
| ".encoder.layers.11.mlp.fc2", | |
| ".encoder.layers.12.attn.qkv", | |
| ".encoder.layers.12.attn.proj", | |
| ".encoder.layers.12.mlp.fc1", | |
| ".encoder.layers.12.mlp.fc2", | |
| ".encoder.layers.13.attn.qkv", | |
| ".encoder.layers.13.attn.proj", | |
| ".encoder.layers.13.mlp.fc1", | |
| ".encoder.layers.13.mlp.fc2", | |
| ".encoder.layers.14.attn.qkv", | |
| ".encoder.layers.14.attn.proj", | |
| ".encoder.layers.14.mlp.fc1", | |
| ".encoder.layers.14.mlp.fc2", | |
| ".encoder.layers.15.attn.qkv", | |
| ".encoder.layers.15.attn.proj", | |
| ".encoder.layers.15.mlp.fc1", | |
| ".encoder.layers.15.mlp.fc2", | |
| ".encoder.layers.16.attn.qkv", | |
| ".encoder.layers.16.attn.proj", | |
| ".encoder.layers.16.mlp.fc1", | |
| ".encoder.layers.16.mlp.fc2", | |
| ".encoder.layers.17.attn.qkv", | |
| ".encoder.layers.17.attn.proj", | |
| ".encoder.layers.17.mlp.fc1", | |
| ".encoder.layers.17.mlp.fc2", | |
| ".encoder.layers.18.attn.qkv", | |
| ".encoder.layers.18.attn.proj", | |
| ".encoder.layers.18.mlp.fc1", | |
| ".encoder.layers.18.mlp.fc2", | |
| ".encoder.layers.19.attn.qkv", | |
| ".encoder.layers.19.attn.proj", | |
| ".encoder.layers.19.mlp.fc1", | |
| ".encoder.layers.19.mlp.fc2", | |
| ".encoder.layers.20.attn.qkv", | |
| ".encoder.layers.20.attn.proj", | |
| ".encoder.layers.20.mlp.fc1", | |
| ".encoder.layers.20.mlp.fc2", | |
| ".encoder.layers.21.attn.qkv", | |
| ".encoder.layers.21.attn.proj", | |
| ".encoder.layers.21.mlp.fc1", | |
| ".encoder.layers.21.mlp.fc2", | |
| ".encoder.layers.22.attn.qkv", | |
| ".encoder.layers.22.attn.proj", | |
| ".encoder.layers.22.mlp.fc1", | |
| ".encoder.layers.22.mlp.fc2", | |
| ".encoder.layers.23.attn.qkv", | |
| ".encoder.layers.23.attn.proj", | |
| ".encoder.layers.23.mlp.fc1", | |
| ".encoder.layers.23.mlp.fc2", | |
| "language_model.lm_head", | |
| "language_model.model.layers.0.mlp.router.query_key_value", | |
| "language_model.model.layers.1.mlp.router.query_key_value", | |
| "language_model.model.layers.2.mlp.router.query_key_value", | |
| "language_model.model.layers.3.mlp.router.query_key_value", | |
| "language_model.model.layers.4.mlp.router.query_key_value", | |
| "language_model.model.layers.5.mlp.router.query_key_value", | |
| "language_model.model.layers.6.mlp.router.query_key_value", | |
| "language_model.model.layers.7.mlp.router.query_key_value", | |
| "language_model.model.layers.8.mlp.router.query_key_value", | |
| "language_model.model.layers.9.mlp.router.query_key_value", | |
| "language_model.model.layers.10.mlp.router.query_key_value", | |
| "language_model.model.layers.11.mlp.router.query_key_value", | |
| "language_model.model.layers.12.mlp.router.query_key_value", | |
| "language_model.model.layers.13.mlp.router.query_key_value", | |
| "language_model.model.layers.14.mlp.router.query_key_value", | |
| "language_model.model.layers.15.mlp.router.query_key_value", | |
| "language_model.model.layers.16.mlp.router.query_key_value", | |
| "language_model.model.layers.17.mlp.router.query_key_value", | |
| "language_model.model.layers.18.mlp.router.query_key_value", | |
| "language_model.model.layers.19.mlp.router.query_key_value", | |
| "language_model.model.layers.20.mlp.router.query_key_value", | |
| "language_model.model.layers.21.mlp.router.query_key_value", | |
| "language_model.model.layers.22.mlp.router.query_key_value", | |
| "language_model.model.layers.23.mlp.router.query_key_value", | |
| "imagemlp.up_proj", | |
| "imagemlp.gate_proj", | |
| "imagemlp.down_proj" | |
| ], | |
| "kv_cache_scheme": null, | |
| "quant_method": "compressed-tensors", | |
| "quantization_status": "compressed", | |
| "sparsity_config": {}, | |
| "transform_config": {}, | |
| "version": "0.11.0" | |
| }, | |
| "select_layer": -1, | |
| "template": "yuan-chat", | |
| "torch_dtype": "bfloat16", | |
| "transformers_version": null, | |
| "use_backbone_lora": 0, | |
| "use_cache": true, | |
| "use_llm_lora": 0, | |
| "use_thumbnail": true, | |
| "vision_config": { | |
| "architectures": [ | |
| "InternVisionModel" | |
| ], | |
| "attention_dropout": 0.0, | |
| "drop_path_rate": 0.0, | |
| "dropout": 0.0, | |
| "hidden_act": "gelu", | |
| "hidden_size": 1024, | |
| "image_size": 448, | |
| "initializer_factor": 1.0, | |
| "initializer_range": 0.02, | |
| "intermediate_size": 4096, | |
| "layer_norm_eps": 1e-06, | |
| "model_type": "intern_vit_6b", | |
| "norm_type": "layer_norm", | |
| "num_attention_heads": 16, | |
| "num_channels": 3, | |
| "num_hidden_layers": 24, | |
| "patch_size": 14, | |
| "qk_normalization": false, | |
| "qkv_bias": true, | |
| "torch_dtype": "bfloat16", | |
| "use_bfloat16": true, | |
| "use_flash_attn": false | |
| } | |
| } | |