*.7z filter=lfs diff=lfs merge=lfs -text *.arrow filter=lfs diff=lfs merge=lfs -text *.bin filter=lfs diff=lfs merge=lfs -text *.bz2 filter=lfs diff=lfs merge=lfs -text *.ckpt filter=lfs diff=lfs merge=lfs -text *.ftz filter=lfs diff=lfs merge=lfs -text *.gz filter=lfs diff=lfs merge=lfs -text *.h5 filter=lfs diff=lfs merge=lfs -text *.joblib filter=lfs diff=lfs merge=lfs -text *.lfs.* filter=lfs diff=lfs merge=lfs -text *.mlmodel filter=lfs diff=lfs merge=lfs -text *.model filter=lfs diff=lfs merge=lfs -text *.msgpack filter=lfs diff=lfs merge=lfs -text *.npy filter=lfs diff=lfs merge=lfs -text *.npz filter=lfs diff=lfs merge=lfs -text *.onnx filter=lfs diff=lfs merge=lfs -text *.ot filter=lfs diff=lfs merge=lfs -text *.parquet filter=lfs diff=lfs merge=lfs -text *.pb filter=lfs diff=lfs merge=lfs -text *.pickle filter=lfs diff=lfs merge=lfs -text *.pkl filter=lfs diff=lfs merge=lfs -text *.pt filter=lfs diff=lfs merge=lfs -text *.pth filter=lfs diff=lfs merge=lfs -text *.rar filter=lfs diff=lfs merge=lfs -text *.safetensors filter=lfs diff=lfs merge=lfs -text saved_model/**/* filter=lfs diff=lfs merge=lfs -text *.tar.* filter=lfs diff=lfs merge=lfs -text *.tar filter=lfs diff=lfs merge=lfs -text *.tflite filter=lfs diff=lfs merge=lfs -text *.tgz filter=lfs diff=lfs merge=lfs -text *.wasm filter=lfs diff=lfs merge=lfs -text *.xz filter=lfs diff=lfs merge=lfs -text *.zip filter=lfs diff=lfs merge=lfs -text *.zst filter=lfs diff=lfs merge=lfs -text *tfevents* filter=lfs diff=lfs merge=lfs -text new-grid/k16_N1024/plots/ablation_conditional_accuracy.png filter=lfs diff=lfs merge=lfs -text new-grid/k16_N1024/plots/ablation_per_position.png filter=lfs diff=lfs merge=lfs -text new-grid/k16_N128/plots/ablation_conditional_accuracy.png filter=lfs diff=lfs merge=lfs -text new-grid/k16_N128/plots/ablation_per_position.png filter=lfs diff=lfs merge=lfs -text new-grid/k16_N256/plots/ablation_conditional_accuracy.png filter=lfs diff=lfs merge=lfs -text new-grid/k16_N256/plots/ablation_per_position.png filter=lfs diff=lfs merge=lfs -text new-grid/k16_N256/plots/length_generalization.png filter=lfs diff=lfs merge=lfs -text new-grid/k16_N512/plots/ablation_conditional_accuracy.png filter=lfs diff=lfs merge=lfs -text new-grid/k16_N512/plots/ablation_per_position.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N1024/plots/ablation_conditional_accuracy.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N1024/plots/ablation_per_position.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N128/plots/ablation_conditional_accuracy.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N128/plots/ablation_per_position.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N128/plots/length_generalization.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N256/plots/ablation_conditional_accuracy.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N256/plots/ablation_per_position.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N256/plots/consecutive_attention.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N256/plots/qk_cross_local.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N256/plots/qk_cross_mean.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N256/plots/qk_cross_overlay.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N256/plots/qk_heatmap.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N256/plots/qk_local_instances.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N256/plots/qk_local_smooth.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N256/plots/qk_monotonicity_summary.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N256/plots/qk_neighbor_cross.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N256/plots/qk_raw_structure.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/ablation_conditional_accuracy.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/ablation_per_position.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/attn1_importance_analysis.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/attn_spread.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/attn_value_distance.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/consecutive_attention.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/consecutive_attention_grid.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/consecutive_attention_grid_ckpt60k.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/intervened_consecutive.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/intervened_consecutive_100k.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/intervened_consecutive_int5.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/l1_qk_interaction.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/l1_vs_l2_qk_comparison.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/length_generalization.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_cross_local.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_cross_mean.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_cross_overlay.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_cross_with_pos.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_deep_decomp.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_fixed_query.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_full_value_path.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_heatmap.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_heatmap_ty.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_heatmap_xt.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_heatmap_xt_60k.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_heatmap_xt_maxscore.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_heatmap_xt_slices.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_heatmap_xt_slices2.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_heatmap_xt_slices3.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_heatmap_xt_split.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_heatmap_xt_split_slices.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_heatmap_zy.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_interaction_decomp.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_local_instances.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_local_smooth.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_monotonicity_summary.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_neighbor_cross.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_neighbor_score.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_query_side_decomp.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_raw_structure.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_residual_path.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_self_score.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_slope_profiles.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_value_heatmap.png filter=lfs diff=lfs merge=lfs -text new-grid/k32_N512/plots/qk_value_heatmap_base.png filter=lfs diff=lfs merge=lfs -text