Investigation 50

Dataset: potions
Model: sonnet4010
Training model: claude-sonnet-4-20250514
Inference model: gpt-4.1-mini
Example count: 10
Patience: 3
Current round: 564
Best round: 251

Rounds

Round	UUID	Started	Completed	Train Acc	Train KT	Val Acc	Val KT	Min Train/Val	Test Acc	Test KT
163	7cdcab7c-b941-43e7-bff9-b547a983016d	2025-06-28	2025-07-02 02:22:06.218687+00:00	0.519	0.285	0.400	0.397	0.400	0.250	0.600
243	3bff6766-8ca5-4faf-aedb-61e934116700	2025-07-02	2025-07-02 02:29:51.353886+00:00	0.631	0.200	0.600	0.222	0.600	0.650	0.188
244	19958f1f-9491-4c58-aa8d-fc7d1d62332e	2025-07-02	2025-07-02 02:43:06.211533+00:00	0.469	0.329	0.550	0.260	0.469	0.300	0.521
245	5d3c35fc-0224-4a42-867d-b1859c86f345	2025-07-02	2025-07-02 02:54:23.339079+00:00	0.500	0.301	0.600	0.222	0.500	0.400	0.397
246	ce29adda-9b39-4b73-95a5-31e32220ec8e	2025-07-02	2025-07-02 03:03:00.908542+00:00	0.731	0.137	0.650	0.188	0.650	0.650	0.188
247	fc7ffe95-848c-40de-91fe-98c8d45403cc	2025-07-02	2025-07-02 03:11:19.010341+00:00	0.556	0.255	0.650	0.188	0.556	0.350	0.455
248	61b61d75-f95b-4567-8801-2082fffb05a4	2025-07-02	2025-07-02 03:20:01.508321+00:00	0.662	0.179	0.700	0.156	0.662	0.500	0.301
249	1635d8bb-ef1e-440b-a032-93c43f0f26d9	2025-07-02	2025-07-02 03:28:19.844679+00:00	0.644	0.192	0.650	0.188	0.644	0.500	0.301
250	0081f066-b650-42f2-a229-b4f118af3441	2025-07-02	2025-07-02 03:36:11.335353+00:00	0.575	0.241	0.650	0.188	0.575	0.350	0.455
251	9149a100-498d-49ba-8db9-381f67008852	2025-07-02	2025-07-02 03:44:27.918266+00:00	0.675	0.171	0.700	0.156	0.675	0.450	0.347
252	9a796e18-f30f-420b-b3da-2e9bb5f49a56	2025-07-02	2025-08-02 08:54:55.816228+00:00	0.675	0.171	0.700	0.156	0.675	0.450	0.347
557	d664c891-71ce-4c6a-a96d-c5aa3630fdd1	2025-08-02	2025-08-02 08:57:33.855276+00:00	0.613	0.213	0.700	0.156	0.613	0.350	0.455
559	47f757f6-360d-479b-8dad-03d4805fb1fb	2025-08-02	2025-08-02 09:15:01.625470+00:00	0.606	0.218	0.550	0.260	0.550	0.450	0.347

Scores

round scores