18 days ago
-
upload: results/part3/raw_generations.json
-
upload: assets/part3/part3_decoding_strategies_comparison.png
-
upload: assets/part3/beam_width_vs_latency.png
-
upload: results/part3/beam_timing_results.json
-
upload: results/part3/decoding_metrics_summary.json
-
upload: configs/part3_decoding.yaml
-
upload: tokenizer/tokenizer_config.json
-
upload: tokenizer/tokenizer.json
-
upload: tokenizer/merges.txt
-
upload: tokenizer/vocab.json
-
upload: assets/part2/part2_best_optimizers_comparison_validation_loss_dynamics.png
-
upload: assets/part2/part2_best_optimizers_comparison_test_bleu_dynamics.png
-
upload: assets/part2/part2_best_optimizers_comparison_dynamics_comparison.png
-
upload: configs/part2_sophia.yaml
-
upload: configs/part2_muon.yaml
-
upload: configs/part2_lion.yaml
-
upload: configs/part2_cautious.yaml
-
upload: configs/part2_adamw.yaml
-
upload: checkpoints/part2/sweep_sophia_lr/sophia_lr_5e-4_dynamics_history.json
-
upload: checkpoints/part2/sweep_sophia_lr/sophia_lr_5e-4_model_best.pt
-
upload: checkpoints/part2/sweep_muon_lr/muon_lr_0.01_dynamics_history.json
-
upload: checkpoints/part2/sweep_muon_lr/muon_lr_0.01_model_best.pt
-
upload: checkpoints/part2/sweep_lion_lr/lion_lr_3e-4_dynamics_history.json
-
upload: checkpoints/part2/sweep_lion_lr/lion_lr_3e-4_model_best.pt
-
upload: checkpoints/part2/sweep_cautious_lr/cautious_lr_1e-3_dynamics_history.json
-
upload: checkpoints/part2/sweep_cautious_lr/cautious_lr_1e-3_model_best.pt
-
upload: checkpoints/part2/sweep_adamw_lr/adamw_lr_1e-3_dynamics_history.json
-
upload: checkpoints/part2/sweep_adamw_lr/adamw_lr_1e-3_model_best.pt
-
upload: assets/part1/part1_variants_loss_ppl_dynamics.png
-
upload: assets/part1/part1_variants_metrics_comparison.png
-
upload: assets/part1/variant3_expert_heatmap.png
-
upload: assets/part1/variant5_loss_curves.png
-
upload: assets/part1/variant4_expert_heatmap.png
-
upload: assets/part1/variant4_loss_curves.png
-
upload: assets/part1/part1_all_expert_heatmaps_grid.png
-
upload: assets/part1/variant5_expert_heatmap.png
-
upload: assets/part1/variant2_loss_curves.png
-
upload: assets/part1/variant3_loss_curves.png
-
upload: assets/part1/variant2_expert_heatmap.png
-
upload: assets/part1/variant1_loss_curves.png
-
upload: results/part1/test_generations_samples.md
-
upload: results/part1/bleu_scores.json
-
upload: results/part1/benchmark_results.json
-
upload: configs/part1_variant5.yaml
-
upload: configs/part1_variant4.yaml
-
upload: configs/part1_variant3.yaml
-
upload: configs/part1_variant2.yaml
-
upload: configs/part1_variant1.yaml
-
upload: checkpoints/part1/variant5/training_history.json
-
upload: checkpoints/part1/variant5/model_variant5_best.pt