| { |
| "architectures": [ |
| "BagelForConditionalGeneration" |
| ], |
| "connector_act": "gelu_pytorch_tanh", |
| "dtype": "bfloat16", |
| "interpolate_pos": false, |
| "latent_patch_size": 2, |
| "llm_config": { |
| "_name_or_path": "", |
| "architectures": [ |
| "Qwen2ForCausalLM" |
| ], |
| "attention_dropout": 0.0, |
| "bos_token_id": 151643, |
| "chunk_size_feed_forward": 0, |
| "dtype": "bfloat16", |
| "eos_token_id": 151645, |
| "hidden_act": "silu", |
| "hidden_size": 3584, |
| "id2label": { |
| "0": "LABEL_0", |
| "1": "LABEL_1" |
| }, |
| "initializer_range": 0.02, |
| "intermediate_size": 18944, |
| "is_encoder_decoder": false, |
| "label2id": { |
| "LABEL_0": 0, |
| "LABEL_1": 1 |
| }, |
| "layer_types": [ |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention", |
| "full_attention" |
| ], |
| "max_position_embeddings": 32768, |
| "max_window_layers": 28, |
| "model_type": "qwen2", |
| "num_attention_heads": 28, |
| "num_hidden_layers": 28, |
| "num_key_value_heads": 4, |
| "output_attentions": false, |
| "output_hidden_states": false, |
| "pad_token_id": null, |
| "problem_type": null, |
| "qk_norm": true, |
| "return_dict": true, |
| "rms_norm_eps": 1e-06, |
| "rope_parameters": { |
| "rope_theta": 1000000.0, |
| "rope_type": "default" |
| }, |
| "sliding_window": null, |
| "tie_word_embeddings": false, |
| "transformers_version": "5.3.0", |
| "use_cache": true, |
| "use_sliding_window": false, |
| "vocab_size": 152064 |
| }, |
| "max_latent_size": 32, |
| "model_type": "bagel", |
| "quantization_config": { |
| "autoround_version": "0.12.0", |
| "bits": 4, |
| "block_name_to_quantize": "language_model.model.layers", |
| "data_type": "int", |
| "extra_config": { |
| ".*moe_gen.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*self_attn\\.k_proj.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*self_attn\\.o_proj.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*self_attn\\.q_proj.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| ".*self_attn\\.v_proj.*": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.0.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.0.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.0.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.0.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.0.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.0.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.0.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.0.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.0.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.0.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.0.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.1.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.1.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.1.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.1.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.1.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.1.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.1.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.1.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.1.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.1.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.1.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.10.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.10.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.10.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.10.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.10.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.10.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.10.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.10.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.10.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.10.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.10.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.11.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.11.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.11.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.11.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.11.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.11.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.11.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.11.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.11.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.11.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.11.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.12.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.12.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.12.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.12.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.12.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.12.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.12.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.12.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.12.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.12.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.12.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.13.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.13.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.13.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.13.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.13.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.13.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.13.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.13.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.13.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.13.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.13.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.14.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.14.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.14.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.14.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.14.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.14.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.14.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.14.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.14.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.14.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.14.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.15.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.15.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.15.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.15.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.15.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.15.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.15.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.15.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.15.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.15.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.15.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.16.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.16.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.16.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.16.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.16.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.16.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.16.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.16.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.16.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.16.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.16.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.17.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.17.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.17.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.17.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.17.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.17.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.17.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.17.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.17.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.17.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.17.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.18.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.18.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.18.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.18.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.18.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.18.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.18.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.18.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.18.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.18.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.18.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.19.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.19.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.19.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.19.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.19.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.19.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.19.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.19.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.19.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.19.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.19.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.2.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.2.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.2.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.2.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.2.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.2.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.2.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.2.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.2.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.2.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.2.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.20.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.20.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.20.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.20.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.20.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.20.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.20.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.20.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.20.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.20.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.20.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.21.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.21.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.21.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.21.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.21.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.21.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.21.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.21.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.21.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.21.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.21.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.22.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.22.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.22.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.22.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.22.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.22.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.22.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.22.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.22.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.22.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.22.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.23.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.23.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.23.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.23.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.23.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.23.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.23.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.23.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.23.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.23.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.23.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.24.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.24.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.24.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.24.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.24.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.24.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.24.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.24.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.24.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.24.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.24.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.25.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.25.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.25.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.25.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.25.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.25.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.25.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.25.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.25.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.25.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.25.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.26.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.26.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.26.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.26.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.26.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.26.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.26.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.26.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.26.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.26.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.26.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.27.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.27.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.27.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.27.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.27.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.27.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.27.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.27.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.27.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.27.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.27.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.3.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.3.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.3.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.3.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.3.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.3.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.3.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.3.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.3.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.3.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.3.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.4.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.4.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.4.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.4.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.4.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.4.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.4.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.4.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.4.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.4.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.4.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.5.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.5.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.5.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.5.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.5.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.5.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.5.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.5.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.5.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.5.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.5.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.6.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.6.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.6.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.6.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.6.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.6.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.6.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.6.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.6.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.6.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.6.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.7.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.7.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.7.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.7.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.7.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.7.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.7.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.7.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.7.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.7.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.7.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.8.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.8.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.8.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.8.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.8.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.8.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.8.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.8.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.8.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.8.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.8.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.9.mlp_moe_gen.down_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.9.mlp_moe_gen.gate_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.9.mlp_moe_gen.up_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.9.self_attn.k_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.9.self_attn.k_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.9.self_attn.o_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.9.self_attn.o_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.9.self_attn.q_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.9.self_attn.q_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.9.self_attn.v_proj": { |
| "bits": 16, |
| "data_type": "float" |
| }, |
| "language_model.model.layers.9.self_attn.v_proj_moe_gen": { |
| "bits": 16, |
| "data_type": "float" |
| } |
| }, |
| "group_size": 128, |
| "packing_format": "auto_round:auto_gptq", |
| "quant_method": "auto-round", |
| "sym": true, |
| "transform_config": {} |
| }, |
| "timestep_shift": 1.0, |
| "transformers_version": "5.3.0", |
| "vae_config": { |
| "downsample": 8, |
| "z_channels": 16 |
| }, |
| "visual_gen": true, |
| "visual_und": true, |
| "vit_config": { |
| "hidden_size": 1152, |
| "image_size": 980, |
| "intermediate_size": 4304, |
| "model_type": "siglip_vision_model", |
| "num_attention_heads": 16, |
| "num_channels": 3, |
| "num_hidden_layers": 27, |
| "patch_size": 14 |
| }, |
| "vit_max_num_patch_per_side": 70 |
| } |
|
|
