| { |
| "architectures": [ |
| "Ovis2_6_MoeForCausalLM" |
| ], |
| "auto_map": { |
| "AutoConfig": "configuration_ovis2_6.Ovis2_6_Moe_Config", |
| "AutoModelForCausalLM": "modeling_ovis2_6.Ovis2_6_MoeForCausalLM" |
| }, |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "bos_token_id": 151643, |
| "conversation_formatter_class": "Qwen3ConversationFormatter", |
| "decoder_sparse_step": 1, |
| "eos_token_id": 151645, |
| "head_dim": 128, |
| "hidden_act": "silu", |
| "hidden_size": 2048, |
| "initializer_range": 0.02, |
| "intermediate_size": 6144, |
| "llm_config": { |
| "_attn_implementation_autoset": true, |
| "_name_or_path": "Qwen/Qwen3-30B-A3B-Thinking-2507", |
| "architectures": [ |
| "Qwen3MoeForCausalLM" |
| ], |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "bos_token_id": 151643, |
| "decoder_sparse_step": 1, |
| "eos_token_id": 151645, |
| "head_dim": 128, |
| "hidden_act": "silu", |
| "hidden_size": 2048, |
| "initializer_range": 0.02, |
| "intermediate_size": 6144, |
| "max_position_embeddings": 262144, |
| "max_window_layers": 48, |
| "mlp_only_layers": [], |
| "model_type": "qwen3_moe", |
| "moe_intermediate_size": 768, |
| "norm_topk_prob": true, |
| "num_attention_heads": 32, |
| "num_experts": 128, |
| "num_experts_per_tok": 8, |
| "num_hidden_layers": 48, |
| "num_key_value_heads": 4, |
| "output_router_logits": false, |
| "rms_norm_eps": 1e-06, |
| "rope_scaling": null, |
| "rope_theta": 10000000, |
| "router_aux_loss_coef": 0.001, |
| "sliding_window": null, |
| "torch_dtype": "bfloat16", |
| "use_cache": false, |
| "use_sliding_window": false, |
| "vocab_size": 151936 |
| }, |
| "max_position_embeddings": 262144, |
| "max_window_layers": 48, |
| "mlp_only_layers": [], |
| "model_type": "ovis2_6_moe", |
| "moe_intermediate_size": 768, |
| "norm_topk_prob": true, |
| "num_attention_heads": 32, |
| "num_experts": 128, |
| "num_experts_per_tok": 8, |
| "num_hidden_layers": 48, |
| "num_key_value_heads": 4, |
| "output_router_logits": false, |
| "rms_norm_eps": 1e-06, |
| "rope_scaling": null, |
| "rope_theta": 10000000, |
| "router_aux_loss_coef": 0.001, |
| "sliding_window": null, |
| "tie_word_embeddings": false, |
| "torch_dtype": "bfloat16", |
| "transformers_version": "4.51.3", |
| "use_cache": false, |
| "use_sliding_window": false, |
| "visual_vocab_size": 65536, |
| "vit_config": { |
| "_attn_implementation_autoset": true, |
| "_name_or_path": "google/siglip2-so400m-patch16-512", |
| "attention_dropout": 0.0, |
| "fullatt_block_indexes": null, |
| "hidden_act": "gelu_pytorch_tanh", |
| "hidden_size": 1152, |
| "hidden_stride": 2, |
| "image_size": 512, |
| "intermediate_size": 4304, |
| "layer_norm_eps": 1e-06, |
| "model_type": "siglip2_navit", |
| "num_attention_heads": 16, |
| "num_channels": 3, |
| "num_hidden_layers": 27, |
| "num_patches": -1, |
| "patch_size": 16, |
| "preserve_original_pe": true, |
| "temporal_patch_size": 1, |
| "torch_dtype": "bfloat16", |
| "use_rope": true, |
| "window_size": 112 |
| }, |
| "vocab_size": 151936 |
| } |
|
|