Files
hamori/checkpoints/finetuned.log.csv
T
H1K0 d9585ec008 data: add fine-tuning run results (lr=3e-5, 30 epochs)
val loss 1.19 → 0.77, val perplexity 3.29 → 2.15.
Best epoch 20, early stop at epoch 30 (patience=10).
Improvement over previous lr=1e-5 run (best val ppl 2.22).

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
2026-05-21 20:52:39 +03:00

32 lines
1.3 KiB
CSV

epoch,train_loss,val_loss,val_ppl,lr,elapsed_s
1,1.173289,1.190454,3.29,2.400000e-05,11.1
2,1.031668,1.017097,2.77,2.998248e-05,9.3
3,0.900459,0.914644,2.50,2.990471e-05,8.2
4,0.826531,0.877679,2.41,2.976507e-05,9.1
5,0.797045,0.851717,2.34,2.956413e-05,9.2
6,0.759114,0.836767,2.31,2.930272e-05,8.7
7,0.736987,0.819369,2.27,2.898194e-05,8.6
8,0.722803,0.806593,2.24,2.860312e-05,9.1
9,0.693306,0.797257,2.22,2.816782e-05,9.1
10,0.683278,0.792332,2.21,2.767785e-05,8.2
11,0.673061,0.787863,2.20,2.713525e-05,9.1
12,0.655914,0.782984,2.19,2.654228e-05,9.1
13,0.643172,0.777573,2.18,2.590139e-05,8.3
14,0.635985,0.774572,2.17,2.521524e-05,9.0
15,0.630730,0.773065,2.17,2.448668e-05,9.1
16,0.622494,0.771514,2.16,2.371874e-05,9.0
17,0.606942,0.769548,2.16,2.291460e-05,8.1
18,0.601119,0.768194,2.16,2.207761e-05,9.1
19,0.601939,0.768208,2.16,2.121123e-05,9.1
20,0.580447,0.766817,2.15,2.031907e-05,8.2
21,0.574881,0.767509,2.15,1.940483e-05,9.1
22,0.576981,0.769625,2.16,1.847230e-05,9.1
23,0.567170,0.770998,2.16,1.752536e-05,8.6
24,0.564600,0.771246,2.16,1.656793e-05,8.7
25,0.556949,0.772251,2.16,1.560399e-05,9.1
26,0.556080,0.770962,2.16,1.463754e-05,9.0
27,0.551530,0.769089,2.16,1.367260e-05,8.2
28,0.542789,0.768548,2.16,1.271317e-05,9.0
29,0.542809,0.770213,2.16,1.176324e-05,9.1
30,0.537889,0.771124,2.16,1.082674e-05,8.2