diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_D.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_D.png new file mode 100644 index 0000000..31e538e Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_D.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_ERG_gap.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_ERG_gap.png new file mode 100644 index 0000000..1d4cfed Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_ERG_gap.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_alpha_clip_xmax.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_alpha_clip_xmax.png new file mode 100644 index 0000000..a7d4af0 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_alpha_clip_xmax.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_alpha_clip_xmax_zoomed.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_alpha_clip_xmax_zoomed.png new file mode 100644 index 0000000..336d65e Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_alpha_clip_xmax_zoomed.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_alpha_raw.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_alpha_raw.png new file mode 100644 index 0000000..a98eff2 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_alpha_raw.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_alpha_raw_zoomed.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_alpha_raw_zoomed.png new file mode 100644 index 0000000..b530c01 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_alpha_raw_zoomed.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_num_traps.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_num_traps.png new file mode 100644 index 0000000..f1a754f Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_num_traps.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_rand_distance.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_rand_distance.png new file mode 100644 index 0000000..76ae388 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/adamw_rand_distance.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/grad_norm_pre_clip.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/grad_norm_pre_clip.png new file mode 100644 index 0000000..0ed574d Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/grad_norm_pre_clip.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_D.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_D.png new file mode 100644 index 0000000..30a6a1c Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_D.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_ERG_gap.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_ERG_gap.png new file mode 100644 index 0000000..717b54c Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_ERG_gap.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_alpha_clip_xmax.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_alpha_clip_xmax.png new file mode 100644 index 0000000..0c0e1ec Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_alpha_clip_xmax.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_alpha_clip_xmax_zoomed.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_alpha_clip_xmax_zoomed.png new file mode 100644 index 0000000..9e45e3b Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_alpha_clip_xmax_zoomed.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_alpha_raw.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_alpha_raw.png new file mode 100644 index 0000000..ecd34e9 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_alpha_raw.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_alpha_raw_zoomed.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_alpha_raw_zoomed.png new file mode 100644 index 0000000..6e0acde Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_alpha_raw_zoomed.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_num_traps.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_num_traps.png new file mode 100644 index 0000000..0896d61 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_num_traps.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_rand_distance.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_rand_distance.png new file mode 100644 index 0000000..2c6e063 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/muon_clip_rand_distance.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/train_loss.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/train_loss.png new file mode 100644 index 0000000..6a040c8 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/train_loss.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/update_to_weight_ratio.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/update_to_weight_ratio.png new file mode 100644 index 0000000..b64730e Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/update_to_weight_ratio.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/val_accuracy.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/val_accuracy.png new file mode 100644 index 0000000..e352b12 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/val_accuracy.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/val_generalization_gap.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/val_generalization_gap.png new file mode 100644 index 0000000..46f9e26 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/val_generalization_gap.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/val_loss.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/val_loss.png new file mode 100644 index 0000000..93fa292 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/val_loss.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/val_perplexity.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/val_perplexity.png new file mode 100644 index 0000000..8a65acb Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/val_perplexity.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/weight_norm.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/weight_norm.png new file mode 100644 index 0000000..edfd2e8 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/four_head_5seed/plots/weight_norm.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/adamw_alpha_raw_vs_clip_xmax.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/adamw_alpha_raw_vs_clip_xmax.png new file mode 100644 index 0000000..9289aea Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/adamw_alpha_raw_vs_clip_xmax.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/adamw_erg_gap_num_traps.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/adamw_erg_gap_num_traps.png new file mode 100644 index 0000000..0f24727 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/adamw_erg_gap_num_traps.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/adamw_performance.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/adamw_performance.png new file mode 100644 index 0000000..1b1f508 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/adamw_performance.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/muon_clip_alpha_raw_vs_clip_xmax.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/muon_clip_alpha_raw_vs_clip_xmax.png new file mode 100644 index 0000000..95c48a5 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/muon_clip_alpha_raw_vs_clip_xmax.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/muon_clip_erg_gap_num_traps.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/muon_clip_erg_gap_num_traps.png new file mode 100644 index 0000000..3ac9cf3 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/muon_clip_erg_gap_num_traps.png differ diff --git a/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/muon_clip_performance.png b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/muon_clip_performance.png new file mode 100644 index 0000000..70f1115 Binary files /dev/null and b/baseline/experiments/nanogpt_one_head_2026_08_21_baseline/results/one_head_5seed/plots/muon_clip_performance.png differ