| { |
| "architectures": [ |
| "Qwen3_5ForConditionalGeneration" |
| ], |
| "image_token_id": 248056, |
| "language_model_only": false, |
| "model_type": "qwen3_5", |
| "text_config": { |
| "attention_bias": false, |
| "attention_dropout": 0.0, |
| "attn_output_gate": true, |
| "bos_token_id": 248044, |
| "dtype": "bfloat16", |
| "eos_token_id": 248044, |
| "full_attention_interval": 4, |
| "head_dim": 256, |
| "hidden_act": "silu", |
| "hidden_size": 5120, |
| "initializer_range": 0.02, |
| "intermediate_size": 17408, |
| "layer_types": [ |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention", |
| "linear_attention", |
| "linear_attention", |
| "linear_attention", |
| "full_attention" |
| ], |
| "linear_conv_kernel_dim": 4, |
| "linear_key_head_dim": 128, |
| "linear_num_key_heads": 16, |
| "linear_num_value_heads": 48, |
| "linear_value_head_dim": 128, |
| "mamba_ssm_dtype": "float32", |
| "max_position_embeddings": 262144, |
| "model_type": "qwen3_5_text", |
| "mtp_num_hidden_layers": 1, |
| "mtp_use_dedicated_embeddings": false, |
| "num_attention_heads": 24, |
| "num_hidden_layers": 64, |
| "num_key_value_heads": 4, |
| "output_gate_type": "swish", |
| "pad_token_id": null, |
| "partial_rotary_factor": 0.25, |
| "rms_norm_eps": 1e-06, |
| "rope_parameters": { |
| "mrope_interleaved": true, |
| "mrope_section": [ |
| 11, |
| 11, |
| 10 |
| ], |
| "partial_rotary_factor": 0.25, |
| "rope_theta": 10000000, |
| "rope_type": "default" |
| }, |
| "tie_word_embeddings": false, |
| "use_cache": true, |
| "vocab_size": 248320 |
| }, |
| "tie_word_embeddings": false, |
| "transformers_version": "4.57.1", |
| "video_token_id": 248057, |
| "vision_config": { |
| "deepstack_visual_indexes": [], |
| "depth": 27, |
| "hidden_act": "gelu_pytorch_tanh", |
| "hidden_size": 1152, |
| "in_channels": 3, |
| "initializer_range": 0.02, |
| "intermediate_size": 4304, |
| "model_type": "qwen3_5", |
| "num_heads": 16, |
| "num_position_embeddings": 2304, |
| "out_hidden_size": 5120, |
| "patch_size": 16, |
| "spatial_merge_size": 2, |
| "temporal_patch_size": 2 |
| }, |
| "vision_end_token_id": 248054, |
| "vision_start_token_id": 248053, |
| "quantization_config": { |
| "config_groups": { |
| "group_0": { |
| "format": "rfp458-pack-quantized", |
| "input_activations": null, |
| "output_activations": null, |
| "targets": [ |
| "Linear" |
| ], |
| "weights": { |
| "num_bits": 4, |
| "type": "float", |
| "group_size": 16, |
| "grid": "iq4_nl", |
| "codebook": [ |
| -127, |
| -104, |
| -83, |
| -65, |
| -49, |
| -35, |
| -22, |
| -10, |
| 1, |
| 13, |
| 25, |
| 38, |
| 53, |
| 69, |
| 89, |
| 113 |
| ], |
| "scale_dtype": "torch.int8", |
| "scale_exp_dtype": "torch.int8", |
| "scale_encoding": "block_float_i8m_i8e", |
| "symmetric": false, |
| "rotation": "hadamard16" |
| } |
| } |
| }, |
| "format": "rfp458-pack-quantized", |
| "quant_method": "rfp458", |
| "quantization_status": "compressed", |
| "version": "rfp458-1.0", |
| "ignore": [ |
| "re:.*\\.visual\\..*", |
| "re:^lm_head", |
| "re:.*\\.mtp\\..*", |
| "re:^model\\.language_model\\.layers\\.0\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.0\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.0\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.0\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.1\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.1\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.1\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.1\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.10\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.10\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.10\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.10\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.11\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.11\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.12\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.12\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.12\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.12\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.13\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.13\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.13\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.13\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.14\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.14\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.14\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.14\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.15\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.15\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.16\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.16\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.16\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.16\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.17\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.17\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.17\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.17\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.18\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.18\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.18\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.18\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.19\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.19\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.2\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.2\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.2\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.2\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.20\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.20\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.20\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.20\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.21\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.21\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.21\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.21\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.22\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.22\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.22\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.22\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.23\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.23\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.24\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.24\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.24\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.24\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.25\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.25\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.25\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.25\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.26\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.26\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.26\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.26\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.27\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.27\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.28\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.28\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.28\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.28\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.29\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.29\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.29\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.29\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.3\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.3\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.30\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.30\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.30\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.30\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.31\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.31\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.32\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.32\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.32\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.32\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.33\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.33\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.33\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.33\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.34\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.34\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.34\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.34\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.35\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.35\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.36\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.36\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.36\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.36\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.37\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.37\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.37\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.37\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.38\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.38\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.38\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.38\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.39\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.39\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.4\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.4\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.4\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.4\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.40\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.40\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.40\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.40\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.41\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.41\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.41\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.41\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.42\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.42\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.42\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.42\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.43\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.43\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.44\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.44\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.44\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.44\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.45\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.45\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.45\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.45\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.46\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.46\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.46\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.46\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.47\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.47\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.48\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.48\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.48\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.48\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.49\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.49\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.49\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.49\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.5\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.5\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.5\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.5\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.50\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.50\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.50\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.50\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.51\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.51\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.52\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.52\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.52\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.52\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.53\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.53\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.53\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.53\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.54\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.54\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.54\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.54\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.55\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.55\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.56\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.56\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.56\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.56\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.57\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.57\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.57\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.57\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.58\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.58\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.58\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.58\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.59\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.59\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.6\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.6\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.6\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.6\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.60\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.60\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.60\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.60\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.61\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.61\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.61\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.61\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.62\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.62\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.62\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.62\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.63\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.63\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.7\\.self_attn\\.k_norm$", |
| "re:^model\\.language_model\\.layers\\.7\\.self_attn\\.q_norm$", |
| "re:^model\\.language_model\\.layers\\.8\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.8\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.8\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.8\\.linear_attn\\.norm$", |
| "re:^model\\.language_model\\.layers\\.9\\.linear_attn\\.conv1d$", |
| "re:^model\\.language_model\\.layers\\.9\\.linear_attn\\.in_proj_a$", |
| "re:^model\\.language_model\\.layers\\.9\\.linear_attn\\.in_proj_b$", |
| "re:^model\\.language_model\\.layers\\.9\\.linear_attn\\.norm$", |
| "re:^mtp\\.fc$", |
| "re:^mtp\\.layers\\.0\\.input_layernorm$", |
| "re:^mtp\\.layers\\.0\\.mlp\\.down_proj$", |
| "re:^mtp\\.layers\\.0\\.mlp\\.gate_proj$", |
| "re:^mtp\\.layers\\.0\\.mlp\\.up_proj$", |
| "re:^mtp\\.layers\\.0\\.post_attention_layernorm$", |
| "re:^mtp\\.layers\\.0\\.self_attn\\.k_norm$", |
| "re:^mtp\\.layers\\.0\\.self_attn\\.q_norm$", |
| "re:^mtp\\.norm$", |
| "re:^mtp\\.pre_fc_norm_embedding$", |
| "re:^mtp\\.pre_fc_norm_hidden$" |
| ] |
| } |
| } |
