{ "architectures": [ "InklingForConditionalGeneration" ], "model_type": "inkling_mm_model", "eos_token_id": 200006, "text_config": { "model_max_length": 1048576, "torch_dtype": "bfloat16", "hidden_size": 4096, "num_hidden_layers": 42, "vocab_size": 201024, "num_attention_heads": 32, "num_key_value_heads": 8, "head_dim": 128, "d_rel": 16, "rel_extent": 1024, "q_bias": false, "o_bias": false, "log_scaling_n_floor": 128000, "log_scaling_alpha": 0.1, "rms_norm_eps": 1e-06, "use_embed_norm": true, "local_layer_ids": [ 0, 1, 2, 3, 4, 6, 7, 8, 9, 10, 12, 13, 14, 15, 16, 18, 19, 20, 21, 22, 24, 25, 26, 27, 28, 30, 31, 32, 33, 34, 36, 37, 38, 39, 40 ], "dense_mlp_idx": 2, "use_sconv": true, "sconv_kernel_size": 4, "unpadded_vocab_size": 200058, "logits_mup_width_multiplier": 16.0, "final_logit_softcapping": null, "swa_head_dim": 128, "swa_num_attention_heads": 32, "swa_num_key_value_heads": 8, "sliding_window_size": 512, "n_routed_experts": 256, "num_experts_per_tok": 6, "n_shared_experts": 2, "shared_expert_sink": true, "dense_intermediate_size": 16384, "intermediate_size": 2048, "route_scale": 8.0, "use_gate_bias": true, "gate_activation": "sigmoid", "norm_after_topk": true, "use_global_scale": true }, "audio_config": { "decoder_dmodel": 4096, "n_mel_bins": 80, "mel_vocab_size": 16, "bias": false, "dmel_min_value": -7.0, "dmel_max_value": 2.0, "use_audio_norm": true, "audio_mode": "dmel" }, "vision_config": { "vision_encoder_type": "hmlp", "decoder_dmodel": 4096, "patch_size": 40, "temporal_patch_size": 2, "n_channels": 3, "n_layers": 4, "use_vision_norm": true }, "mtp_config": { "num_nextn_predict_layers": 8, "chain_hidden_post_norm": false, "local_layer_ids": [ 0, 2, 4, 5, 6, 7 ] }, "quantization_config": { "quant_method": "auto-round", "packing_format": "auto_round:llm_compressor", "bits": 4, "group_size": 32, "sym": true, "data_type": "mx_fp", "iters": 0, "model_free": true, "autoround_version": "0.15.0", "enable_quanted_input": false, "block_name_to_quantize": "model.llm.layers", "act_bits": 16, "act_data_type": "fp", "act_dynamic": true, "act_group_size": 128, "act_sym": true, "extra_config": { "mlp.experts": { "act_bits": 4, "act_group_size": 32, "act_data_type": "mx_fp" }, "model.audio.encoder": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.13.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.3.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.34.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.37.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.37.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.11.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.13.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.18.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.19.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.2.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.2.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.20.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.21.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.22.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.23.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.25.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.27.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.29.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.3.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.3.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.32.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.32.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.37.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.4.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.4.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.40.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.15.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.23.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.30.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.34.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.36.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.4.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.6.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.8.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.9.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.18.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.27.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.30.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.0.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.0.mlp.w13_dn": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.14.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.20.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.23.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.23.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.23.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.24.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.26.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.27.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.35.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.36.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.37.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.41.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.5.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.7.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.visual.layers.linear_2": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.12.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.14.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.28.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.31.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.35.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.38.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.8.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.9.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.11.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.12.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.14.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.15.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.19.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.2.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.24.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.28.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.3.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.33.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.34.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.35.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.6.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.0.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.10.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.13.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.15.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.17.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.20.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.36.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.38.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.41.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.9.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.9.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.0.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.10.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.10.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.19.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.21.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.28.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.30.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.35.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.40.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.8.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.0.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.0.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.1.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.1.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.16.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.21.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.21.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.23.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.24.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.25.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.26.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.27.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.29.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.33.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.36.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.39.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.13.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.15.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.16.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.18.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.20.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.22.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.27.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.29.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.16.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.17.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.19.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.27.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.29.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.38.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.39.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.4.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.14.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.16.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.18.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.22.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.28.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.31.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.40.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.5.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.7.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.8.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.visual.layers.linear_1": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.visual.layers.linear_3": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.0.mlp.w2_md": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.18.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.20.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.24.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.25.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.3.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.33.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.33.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.39.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.12.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.13.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.16.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.21.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.28.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.29.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.34.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.38.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.38.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.4.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.4.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.7.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.1.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.11.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.16.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.18.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.26.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.26.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.31.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.36.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.37.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.39.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.40.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.40.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.5.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.8.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.9.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.1.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.1.mlp.w2_md": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.10.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.17.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.22.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.38.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.41.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.41.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.8.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.visual.layers.linear_0": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.31.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.31.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.32.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.1.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.20.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.24.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.25.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.17.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.17.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.22.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.1.mlp.w13_dn": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.12.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.25.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.35.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.36.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.10.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.14.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.15.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.2.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.24.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.25.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.26.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.29.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.32.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.32.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.35.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.5.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.6.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.6.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.6.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.9.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.10.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.11.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.12.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.21.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.22.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.3.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.6.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.7.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.11.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.12.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.17.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.19.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.19.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.28.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.33.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.33.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.40.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.15.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.30.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.31.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.0.input_proj": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.0.transformer_block.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.0.transformer_block.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.0.transformer_block.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.0.transformer_block.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.0.transformer_block.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.0.transformer_block.mlp.w13_dn": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.0.transformer_block.mlp.w2_md": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.1.input_proj": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.1.transformer_block.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.1.transformer_block.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.1.transformer_block.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.1.transformer_block.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.1.transformer_block.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.1.transformer_block.mlp.w13_dn": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.1.transformer_block.mlp.w2_md": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.2.input_proj": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.2.transformer_block.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.2.transformer_block.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.2.transformer_block.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.2.transformer_block.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.2.transformer_block.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.2.transformer_block.mlp.w13_dn": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.2.transformer_block.mlp.w2_md": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.3.input_proj": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.3.transformer_block.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.3.transformer_block.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.3.transformer_block.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.3.transformer_block.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.3.transformer_block.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.3.transformer_block.mlp.w13_dn": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.3.transformer_block.mlp.w2_md": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.4.input_proj": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.4.transformer_block.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.4.transformer_block.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.4.transformer_block.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.4.transformer_block.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.4.transformer_block.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.4.transformer_block.mlp.w13_dn": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.4.transformer_block.mlp.w2_md": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.5.input_proj": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.5.transformer_block.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.5.transformer_block.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.5.transformer_block.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.5.transformer_block.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.5.transformer_block.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.5.transformer_block.mlp.w13_dn": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.5.transformer_block.mlp.w2_md": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.6.input_proj": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.6.transformer_block.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.6.transformer_block.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.6.transformer_block.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.6.transformer_block.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.6.transformer_block.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.6.transformer_block.mlp.w13_dn": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.6.transformer_block.mlp.w2_md": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.7.input_proj": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.7.transformer_block.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.7.transformer_block.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.7.transformer_block.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.7.transformer_block.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.7.transformer_block.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.7.transformer_block.mlp.w13_dn": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.mtp.layers.7.transformer_block.mlp.w2_md": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.14.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.2.attn.wr_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.30.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.39.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.39.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.7.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.13.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.30.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.34.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.41.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.5.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.7.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.11.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.37.attn.wv_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.41.attn.wk_dv": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.26.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.32.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.5.attn.wo_ud": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.2.attn.wq_du": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" }, "model.llm.layers.34.mlp.gate": { "bits": 16, "data_type": "float", "act_bits": 16, "act_data_type": "float" } } } }