End of training

README.md

@@ -18,8 +18,8 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [microsoft/deberta-v3-base](https://huggingface.co/microsoft/deberta-v3-base) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.2219
-- F1: 0.5208
+- Loss: 1.0544
+- F1: 0.5493
 
 ## Model description
 

all_results.json

@@ -1,15 +1,15 @@
 {
-    "epoch": 5.0,
-    "eval_f1": 0.531250378962142,
-    "eval_loss": 1.0681172609329224,
-    "eval_runtime": 2.6267,
+    "epoch": 10.0,
+    "eval_f1": 0.5492932407699946,
+    "eval_loss": 1.054356575012207,
+    "eval_runtime": 2.6206,
     "eval_samples": 1101,
-    "eval_samples_per_second": 419.155,
-    "eval_steps_per_second": 6.853,
-    "total_flos": 6586294115808000.0,
-    "train_loss": 0.9968819860202163,
-    "train_runtime": 296.7277,
+    "eval_samples_per_second": 420.131,
+    "eval_steps_per_second": 6.869,
+    "total_flos": 1.3172588231616e+16,
+    "train_loss": 0.8062449658094947,
+    "train_runtime": 588.6173,
     "train_samples": 8544,
-    "train_samples_per_second": 143.97,
-    "train_steps_per_second": 2.258
+    "train_samples_per_second": 145.154,
+    "train_steps_per_second": 2.277
 }

eval_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 5.0,
-    "eval_f1": 0.531250378962142,
-    "eval_loss": 1.0681172609329224,
-    "eval_runtime": 2.6267,
+    "epoch": 10.0,
+    "eval_f1": 0.5492932407699946,
+    "eval_loss": 1.054356575012207,
+    "eval_runtime": 2.6206,
     "eval_samples": 1101,
-    "eval_samples_per_second": 419.155,
-    "eval_steps_per_second": 6.853
+    "eval_samples_per_second": 420.131,
+    "eval_steps_per_second": 6.869
 }

predict_results.txt

@@ -10,13 +10,13 @@ index	prediction
 8	2
 9	3
 10	2
-11	1
+11	2
 12	0
 13	3
 14	0
 15	2
 16	3
-17	1
+17	2
 18	2
 19	4
 20	3
@@ -25,15 +25,15 @@ index	prediction
 23	3
 24	2
 25	1
-26	2
+26	3
 27	2
 28	3
 29	3
-30	1
+30	2
 31	1
 32	2
 33	1
-34	4
+34	3
 35	1
 36	3
 37	3
@@ -49,7 +49,7 @@ index	prediction
 47	1
 48	3
 49	3
-50	3
+50	2
 51	1
 52	3
 53	0
@@ -59,7 +59,7 @@ index	prediction
 57	3
 58	1
 59	3
-60	1
+60	0
 61	4
 62	3
 63	0
@@ -75,14 +75,14 @@ index	prediction
 73	0
 74	3
 75	3
-76	4
+76	3
 77	2
 78	4
 79	0
 80	0
 81	2
 82	2
-83	0
+83	1
 84	3
 85	0
 86	1
@@ -101,7 +101,7 @@ index	prediction
 99	3
 100	3
 101	2
-102	2
+102	3
 103	3
 104	1
 105	0
@@ -134,7 +134,7 @@ index	prediction
 132	4
 133	0
 134	0
-135	2
+135	3
 136	3
 137	2
 138	3
@@ -155,8 +155,8 @@ index	prediction
 153	0
 154	3
 155	0
-156	1
-157	0
+156	2
+157	1
 158	1
 159	3
 160	1
@@ -164,7 +164,7 @@ index	prediction
 162	3
 163	1
 164	3
-165	1
+165	2
 166	1
 167	3
 168	4
@@ -175,22 +175,22 @@ index	prediction
 173	0
 174	3
 175	0
-176	4
+176	3
 177	1
 178	1
 179	0
 180	1
-181	4
+181	3
 182	1
 183	1
 184	0
-185	4
-186	4
+185	3
+186	3
 187	2
 188	3
 189	4
 190	0
-191	4
+191	3
 192	4
 193	4
 194	4
@@ -215,10 +215,10 @@ index	prediction
 213	4
 214	0
 215	2
-216	4
-217	0
+216	3
+217	1
 218	4
-219	2
+219	3
 220	2
 221	3
 222	0
@@ -227,7 +227,7 @@ index	prediction
 225	2
 226	4
 227	0
-228	4
+228	3
 229	3
 230	4
 231	0
@@ -258,10 +258,10 @@ index	prediction
 256	1
 257	0
 258	4
-259	1
+259	2
 260	1
 261	4
-262	4
+262	3
 263	2
 264	4
 265	3
@@ -278,12 +278,12 @@ index	prediction
 276	3
 277	3
 278	4
-279	1
+279	2
 280	4
-281	1
+281	2
 282	4
 283	2
-284	4
+284	3
 285	3
 286	4
 287	0
@@ -307,7 +307,7 @@ index	prediction
 305	0
 306	2
 307	4
-308	4
+308	3
 309	0
 310	4
 311	3
@@ -322,11 +322,11 @@ index	prediction
 320	4
 321	3
 322	2
-323	3
+323	2
 324	1
 325	0
 326	4
-327	4
+327	3
 328	3
 329	4
 330	3
@@ -375,16 +375,16 @@ index	prediction
 373	3
 374	1
 375	3
-376	2
-377	1
+376	3
+377	2
 378	2
 379	2
 380	4
 381	4
 382	3
-383	4
+383	3
 384	3
-385	1
+385	0
 386	2
 387	0
 388	1
@@ -397,7 +397,7 @@ index	prediction
 395	2
 396	4
 397	4
-398	4
+398	3
 399	1
 400	4
 401	1
@@ -409,7 +409,7 @@ index	prediction
 407	2
 408	1
 409	4
-410	1
+410	2
 411	4
 412	1
 413	4
@@ -430,11 +430,11 @@ index	prediction
 428	4
 429	4
 430	2
-431	4
+431	3
 432	0
 433	3
 434	1
-435	0
+435	1
 436	1
 437	1
 438	0
@@ -450,17 +450,17 @@ index	prediction
 448	4
 449	1
 450	2
-451	1
-452	4
+451	2
+452	3
 453	3
 454	4
 455	1
 456	4
 457	0
-458	4
+458	3
 459	2
-460	1
-461	4
+460	2
+461	3
 462	3
 463	0
 464	2
@@ -468,7 +468,7 @@ index	prediction
 466	1
 467	0
 468	1
-469	1
+469	2
 470	3
 471	1
 472	3
@@ -480,7 +480,7 @@ index	prediction
 478	1
 479	1
 480	3
-481	4
+481	3
 482	2
 483	0
 484	0
@@ -502,7 +502,7 @@ index	prediction
 500	1
 501	1
 502	3
-503	1
+503	2
 504	1
 505	0
 506	4
@@ -527,9 +527,9 @@ index	prediction
 525	0
 526	4
 527	3
-528	2
+528	3
 529	3
-530	3
+530	2
 531	1
 532	1
 533	1
@@ -542,20 +542,20 @@ index	prediction
 540	3
 541	4
 542	2
-543	4
+543	3
 544	2
 545	3
 546	1
 547	3
 548	3
 549	0
-550	1
+550	2
 551	1
 552	0
 553	3
-554	1
+554	2
 555	3
-556	4
+556	3
 557	1
 558	2
 559	0
@@ -566,7 +566,7 @@ index	prediction
 564	1
 565	0
 566	3
-567	4
+567	3
 568	2
 569	4
 570	3
@@ -580,7 +580,7 @@ index	prediction
 578	4
 579	1
 580	3
-581	0
+581	1
 582	4
 583	3
 584	0
@@ -594,7 +594,7 @@ index	prediction
 592	2
 593	2
 594	4
-595	1
+595	2
 596	4
 597	3
 598	4
@@ -608,13 +608,13 @@ index	prediction
 606	3
 607	1
 608	4
-609	0
+609	1
 610	1
 611	1
 612	1
 613	3
 614	4
-615	0
+615	1
 616	1
 617	1
 618	1
@@ -623,9 +623,9 @@ index	prediction
 621	3
 622	2
 623	1
-624	2
+624	3
 625	1
-626	1
+626	2
 627	1
 628	0
 629	4
@@ -642,7 +642,7 @@ index	prediction
 640	1
 641	3
 642	4
-643	1
+643	2
 644	1
 645	2
 646	3
@@ -661,7 +661,7 @@ index	prediction
 659	1
 660	3
 661	4
-662	1
+662	2
 663	1
 664	3
 665	1
@@ -669,12 +669,12 @@ index	prediction
 667	3
 668	4
 669	1
-670	4
+670	3
 671	3
 672	1
-673	2
+673	3
 674	3
-675	1
+675	2
 676	4
 677	3
 678	4
@@ -686,7 +686,7 @@ index	prediction
 684	3
 685	2
 686	3
-687	2
+687	3
 688	3
 689	4
 690	4
@@ -699,11 +699,11 @@ index	prediction
 697	4
 698	1
 699	3
-700	3
+700	2
 701	1
 702	1
 703	4
-704	1
+704	2
 705	2
 706	3
 707	0
@@ -714,7 +714,7 @@ index	prediction
 712	1
 713	3
 714	3
-715	1
+715	2
 716	2
 717	1
 718	0
@@ -727,11 +727,11 @@ index	prediction
 725	1
 726	3
 727	0
-728	4
+728	3
 729	0
 730	1
 731	4
-732	2
+732	3
 733	2
 734	4
 735	0
@@ -740,13 +740,13 @@ index	prediction
 738	3
 739	0
 740	3
-741	4
-742	1
+741	3
+742	2
 743	1
 744	4
 745	4
 746	3
-747	1
+747	2
 748	4
 749	3
 750	4
@@ -759,7 +759,7 @@ index	prediction
 757	3
 758	2
 759	1
-760	1
+760	2
 761	2
 762	4
 763	4
@@ -770,8 +770,8 @@ index	prediction
 768	0
 769	0
 770	3
-771	1
-772	1
+771	2
+772	2
 773	1
 774	4
 775	2
@@ -781,7 +781,7 @@ index	prediction
 779	2
 780	0
 781	3
-782	3
+782	2
 783	4
 784	1
 785	3
@@ -795,7 +795,7 @@ index	prediction
 793	4
 794	4
 795	3
-796	4
+796	3
 797	0
 798	1
 799	2
@@ -814,7 +814,7 @@ index	prediction
 812	4
 813	0
 814	1
-815	0
+815	1
 816	2
 817	0
 818	1
@@ -832,13 +832,13 @@ index	prediction
 830	3
 831	4
 832	4
-833	0
+833	1
 834	4
 835	3
 836	4
 837	1
 838	3
-839	4
+839	3
 840	3
 841	3
 842	3
@@ -848,7 +848,7 @@ index	prediction
 846	3
 847	4
 848	0
-849	0
+849	1
 850	4
 851	3
 852	4
@@ -858,7 +858,7 @@ index	prediction
 856	4
 857	2
 858	1
-859	0
+859	1
 860	1
 861	2
 862	1
@@ -868,7 +868,7 @@ index	prediction
 866	4
 867	2
 868	3
-869	1
+869	2
 870	3
 871	2
 872	1
@@ -879,13 +879,13 @@ index	prediction
 877	4
 878	3
 879	3
-880	4
+880	3
 881	3
 882	4
 883	1
 884	3
 885	4
-886	4
+886	3
 887	3
 888	4
 889	2
@@ -905,16 +905,16 @@ index	prediction
 903	3
 904	3
 905	4
-906	1
+906	2
 907	0
 908	4
 909	2
 910	4
 911	2
-912	0
+912	1
 913	1
-914	4
-915	2
+914	3
+915	3
 916	3
 917	3
 918	3
@@ -964,7 +964,7 @@ index	prediction
 962	0
 963	4
 964	1
-965	2
+965	3
 966	1
 967	3
 968	1
@@ -977,7 +977,7 @@ index	prediction
 975	3
 976	1
 977	4
-978	1
+978	2
 979	0
 980	1
 981	1
@@ -988,7 +988,7 @@ index	prediction
 986	0
 987	2
 988	1
-989	2
+989	3
 990	1
 991	0
 992	1
@@ -1013,7 +1013,7 @@ index	prediction
 1011	4
 1012	3
 1013	3
-1014	1
+1014	2
 1015	2
 1016	2
 1017	3
@@ -1036,7 +1036,7 @@ index	prediction
 1034	1
 1035	2
 1036	4
-1037	4
+1037	3
 1038	3
 1039	1
 1040	1
@@ -1046,7 +1046,7 @@ index	prediction
 1044	4
 1045	4
 1046	4
-1047	1
+1047	2
 1048	1
 1049	4
 1050	4
@@ -1103,14 +1103,14 @@ index	prediction
 1101	4
 1102	3
 1103	4
-1104	1
+1104	0
 1105	1
 1106	3
-1107	0
+1107	1
 1108	0
 1109	1
 1110	1
-1111	1
+1111	2
 1112	3
 1113	1
 1114	3
@@ -1150,11 +1150,11 @@ index	prediction
 1148	4
 1149	3
 1150	1
-1151	1
+1151	0
 1152	1
 1153	2
 1154	3
-1155	3
+1155	2
 1156	1
 1157	0
 1158	1
@@ -1167,7 +1167,7 @@ index	prediction
 1165	4
 1166	4
 1167	4
-1168	1
+1168	0
 1169	1
 1170	1
 1171	4
@@ -1190,14 +1190,14 @@ index	prediction
 1188	1
 1189	3
 1190	3
-1191	1
-1192	1
+1191	2
+1192	2
 1193	1
 1194	3
 1195	0
 1196	2
-1197	0
-1198	2
+1197	1
+1198	3
 1199	2
 1200	2
 1201	3
@@ -1205,7 +1205,7 @@ index	prediction
 1203	2
 1204	1
 1205	3
-1206	3
+1206	2
 1207	2
 1208	3
 1209	4
@@ -1215,7 +1215,7 @@ index	prediction
 1213	2
 1214	1
 1215	4
-1216	1
+1216	2
 1217	0
 1218	4
 1219	0
@@ -1236,8 +1236,8 @@ index	prediction
 1234	3
 1235	3
 1236	3
-1237	0
-1238	4
+1237	1
+1238	3
 1239	1
 1240	3
 1241	0
@@ -1264,12 +1264,12 @@ index	prediction
 1262	0
 1263	3
 1264	4
-1265	4
+1265	3
 1266	2
 1267	1
 1268	3
 1269	2
-1270	1
+1270	2
 1271	1
 1272	3
 1273	1
@@ -1301,12 +1301,12 @@ index	prediction
 1299	0
 1300	4
 1301	3
-1302	1
+1302	2
 1303	4
 1304	0
 1305	1
 1306	1
-1307	1
+1307	2
 1308	1
 1309	3
 1310	3
@@ -1318,7 +1318,7 @@ index	prediction
 1316	1
 1317	2
 1318	2
-1319	1
+1319	0
 1320	4
 1321	4
 1322	2
@@ -1336,7 +1336,7 @@ index	prediction
 1334	3
 1335	1
 1336	1
-1337	1
+1337	2
 1338	3
 1339	1
 1340	3
@@ -1347,8 +1347,8 @@ index	prediction
 1345	4
 1346	0
 1347	3
-1348	3
-1349	1
+1348	2
+1349	2
 1350	3
 1351	1
 1352	4
@@ -1379,35 +1379,35 @@ index	prediction
 1377	1
 1378	1
 1379	1
-1380	0
+1380	1
 1381	1
 1382	1
 1383	2
 1384	3
-1385	0
+1385	1
 1386	3
 1387	4
 1388	4
 1389	1
 1390	1
-1391	2
+1391	3
 1392	2
 1393	1
-1394	0
+1394	1
 1395	3
 1396	1
 1397	1
 1398	1
 1399	3
 1400	4
-1401	0
+1401	1
 1402	0
 1403	3
-1404	4
+1404	3
 1405	3
 1406	1
 1407	2
-1408	3
+1408	2
 1409	3
 1410	2
 1411	1
@@ -1440,7 +1440,7 @@ index	prediction
 1438	4
 1439	4
 1440	3
-1441	1
+1441	2
 1442	1
 1443	1
 1444	0
@@ -1455,7 +1455,7 @@ index	prediction
 1453	1
 1454	0
 1455	3
-1456	1
+1456	2
 1457	0
 1458	0
 1459	1
@@ -1499,9 +1499,9 @@ index	prediction
 1497	4
 1498	4
 1499	4
-1500	4
+1500	3
 1501	3
-1502	2
+1502	3
 1503	1
 1504	1
 1505	2
@@ -1518,7 +1518,7 @@ index	prediction
 1516	3
 1517	1
 1518	2
-1519	2
+1519	3
 1520	0
 1521	1
 1522	2
@@ -1530,11 +1530,11 @@ index	prediction
 1528	4
 1529	1
 1530	1
-1531	1
-1532	4
+1531	2
+1532	3
 1533	3
-1534	4
-1535	1
+1534	3
+1535	2
 1536	2
 1537	2
 1538	4
@@ -1542,7 +1542,7 @@ index	prediction
 1540	3
 1541	0
 1542	3
-1543	1
+1543	2
 1544	1
 1545	2
 1546	3
@@ -1558,7 +1558,7 @@ index	prediction
 1556	1
 1557	3
 1558	1
-1559	3
+1559	2
 1560	0
 1561	1
 1562	1
@@ -1577,7 +1577,7 @@ index	prediction
 1575	2
 1576	1
 1577	2
-1578	1
+1578	0
 1579	0
 1580	1
 1581	2
@@ -1589,14 +1589,14 @@ index	prediction
 1587	1
 1588	4
 1589	2
-1590	1
-1591	1
+1590	2
+1591	2
 1592	4
 1593	2
 1594	1
 1595	3
 1596	4
-1597	2
+1597	3
 1598	1
 1599	3
 1600	2
@@ -1606,19 +1606,19 @@ index	prediction
 1604	3
 1605	1
 1606	2
-1607	4
+1607	3
 1608	2
 1609	1
 1610	1
-1611	1
+1611	2
 1612	1
 1613	0
 1614	4
 1615	3
 1616	3
-1617	1
+1617	2
 1618	2
-1619	4
+1619	3
 1620	4
 1621	1
 1622	3
@@ -1633,7 +1633,7 @@ index	prediction
 1631	3
 1632	4
 1633	1
-1634	4
+1634	3
 1635	4
 1636	3
 1637	0
@@ -1672,7 +1672,7 @@ index	prediction
 1670	3
 1671	3
 1672	1
-1673	4
+1673	3
 1674	3
 1675	3
 1676	3
@@ -1688,17 +1688,17 @@ index	prediction
 1686	3
 1687	1
 1688	3
-1689	4
+1689	3
 1690	0
 1691	2
 1692	2
-1693	1
-1694	4
+1693	2
+1694	3
 1695	3
 1696	1
 1697	4
 1698	1
-1699	1
+1699	0
 1700	1
 1701	2
 1702	3
@@ -1715,7 +1715,7 @@ index	prediction
 1713	4
 1714	0
 1715	0
-1716	1
+1716	0
 1717	4
 1718	3
 1719	1
@@ -1723,7 +1723,7 @@ index	prediction
 1721	1
 1722	1
 1723	2
-1724	1
+1724	2
 1725	1
 1726	4
 1727	3
@@ -1740,7 +1740,7 @@ index	prediction
 1738	1
 1739	3
 1740	3
-1741	4
+1741	3
 1742	4
 1743	1
 1744	4
@@ -1755,9 +1755,9 @@ index	prediction
 1753	1
 1754	3
 1755	3
-1756	4
+1756	3
 1757	3
-1758	1
+1758	2
 1759	4
 1760	3
 1761	3
@@ -1778,13 +1778,13 @@ index	prediction
 1776	2
 1777	4
 1778	0
-1779	4
+1779	3
 1780	1
 1781	1
 1782	1
 1783	3
-1784	4
-1785	0
+1784	3
+1785	1
 1786	4
 1787	3
 1788	2
@@ -1793,27 +1793,27 @@ index	prediction
 1791	1
 1792	1
 1793	4
-1794	4
+1794	3
 1795	1
 1796	3
 1797	2
 1798	3
 1799	3
 1800	1
-1801	4
-1802	1
+1801	3
+1802	2
 1803	3
 1804	4
 1805	3
 1806	1
 1807	1
-1808	1
+1808	2
 1809	0
 1810	1
 1811	3
 1812	2
 1813	3
-1814	4
+1814	3
 1815	3
 1816	0
 1817	1
@@ -1882,7 +1882,7 @@ index	prediction
 1880	0
 1881	3
 1882	3
-1883	2
+1883	3
 1884	1
 1885	0
 1886	0
@@ -1899,7 +1899,7 @@ index	prediction
 1897	1
 1898	4
 1899	4
-1900	4
+1900	3
 1901	3
 1902	2
 1903	4
@@ -1922,7 +1922,7 @@ index	prediction
 1920	3
 1921	1
 1922	0
-1923	1
+1923	2
 1924	1
 1925	3
 1926	1
@@ -1939,7 +1939,7 @@ index	prediction
 1937	1
 1938	4
 1939	1
-1940	1
+1940	2
 1941	3
 1942	3
 1943	4
@@ -1953,7 +1953,7 @@ index	prediction
 1951	3
 1952	3
 1953	3
-1954	0
+1954	1
 1955	1
 1956	2
 1957	3
@@ -1967,7 +1967,7 @@ index	prediction
 1965	2
 1966	0
 1967	1
-1968	1
+1968	2
 1969	2
 1970	2
 1971	0
@@ -1980,7 +1980,7 @@ index	prediction
 1978	4
 1979	3
 1980	4
-1981	1
+1981	2
 1982	0
 1983	1
 1984	0
@@ -2028,7 +2028,7 @@ index	prediction
 2026	4
 2027	3
 2028	1
-2029	1
+2029	2
 2030	3
 2031	3
 2032	1
@@ -2039,20 +2039,20 @@ index	prediction
 2037	2
 2038	2
 2039	3
-2040	4
+2040	3
 2041	1
 2042	0
 2043	3
 2044	1
 2045	1
-2046	1
+2046	0
 2047	3
 2048	1
-2049	1
+2049	2
 2050	3
 2051	1
 2052	1
-2053	0
+2053	1
 2054	1
 2055	2
 2056	3
@@ -2065,12 +2065,12 @@ index	prediction
 2063	3
 2064	1
 2065	3
-2066	1
+2066	2
 2067	0
 2068	1
 2069	2
-2070	4
-2071	1
+2070	3
+2071	2
 2072	3
 2073	3
 2074	2
@@ -2079,7 +2079,7 @@ index	prediction
 2077	2
 2078	4
 2079	1
-2080	1
+2080	2
 2081	2
 2082	1
 2083	1
@@ -2096,7 +2096,7 @@ index	prediction
 2094	4
 2095	1
 2096	2
-2097	4
+2097	3
 2098	1
 2099	1
 2100	1
@@ -2139,7 +2139,7 @@ index	prediction
 2137	0
 2138	0
 2139	2
-2140	4
+2140	3
 2141	2
 2142	1
 2143	2
@@ -2149,7 +2149,7 @@ index	prediction
 2147	1
 2148	3
 2149	4
-2150	4
+2150	3
 2151	3
 2152	4
 2153	3
@@ -2158,13 +2158,13 @@ index	prediction
 2156	4
 2157	3
 2158	3
-2159	0
+2159	1
 2160	4
 2161	1
-2162	4
+2162	3
 2163	4
 2164	1
-2165	3
+2165	4
 2166	1
 2167	1
 2168	3
@@ -2183,26 +2183,26 @@ index	prediction
 2181	4
 2182	4
 2183	3
-2184	4
+2184	3
 2185	0
-2186	1
+2186	0
 2187	1
 2188	1
 2189	3
-2190	1
+2190	2
 2191	4
-2192	4
+2192	3
 2193	3
 2194	0
-2195	1
+2195	2
 2196	1
 2197	1
 2198	4
 2199	1
 2200	1
-2201	1
+2201	2
 2202	3
-2203	1
+2203	2
 2204	1
 2205	0
 2206	1

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 5.0,
-    "total_flos": 6586294115808000.0,
-    "train_loss": 0.9968819860202163,
-    "train_runtime": 296.7277,
+    "epoch": 10.0,
+    "total_flos": 1.3172588231616e+16,
+    "train_loss": 0.8062449658094947,
+    "train_runtime": 588.6173,
     "train_samples": 8544,
-    "train_samples_per_second": 143.97,
-    "train_steps_per_second": 2.258
+    "train_samples_per_second": 145.154,
+    "train_steps_per_second": 2.277
 }

trainer_state.json

@@ -1,550 +1,1082 @@
 {
-  "best_metric": 0.531250378962142,
-  "best_model_checkpoint": "/scratch/06782/ysu707/results/sst5/deberta_base/checkpoint-600",
-  "epoch": 5.0,
+  "best_metric": 0.5492932407699946,
+  "best_model_checkpoint": "/scratch/06782/ysu707/results/sst5/deberta_base/checkpoint-500",
+  "epoch": 10.0,
   "eval_steps": 100,
-  "global_step": 670,
+  "global_step": 1340,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.07462686567164178,
-      "grad_norm": 0.7592889666557312,
-      "learning_rate": 9.95475113122172e-06,
-      "loss": 1.6025,
+      "grad_norm": 0.7684511542320251,
+      "learning_rate": 7.1428571428571436e-06,
+      "loss": 1.6032,
       "step": 10
     },
     {
       "epoch": 0.14925373134328357,
-      "grad_norm": 1.0113656520843506,
-      "learning_rate": 9.803921568627451e-06,
-      "loss": 1.5865,
+      "grad_norm": 1.0371614694595337,
+      "learning_rate": 9.95475113122172e-06,
+      "loss": 1.5902,
       "step": 20
     },
     {
       "epoch": 0.22388059701492538,
-      "grad_norm": 0.9652906060218811,
-      "learning_rate": 9.653092006033184e-06,
-      "loss": 1.585,
+      "grad_norm": 0.9901672005653381,
+      "learning_rate": 9.879336349924586e-06,
+      "loss": 1.5876,
       "step": 30
     },
     {
       "epoch": 0.29850746268656714,
-      "grad_norm": 1.4452743530273438,
-      "learning_rate": 9.502262443438914e-06,
-      "loss": 1.5718,
+      "grad_norm": 1.3719103336334229,
+      "learning_rate": 9.803921568627451e-06,
+      "loss": 1.5746,
       "step": 40
     },
     {
       "epoch": 0.373134328358209,
-      "grad_norm": 0.9761541485786438,
-      "learning_rate": 9.351432880844647e-06,
+      "grad_norm": 0.9476407170295715,
+      "learning_rate": 9.728506787330318e-06,
       "loss": 1.5571,
       "step": 50
     },
     {
       "epoch": 0.44776119402985076,
-      "grad_norm": 1.4800972938537598,
-      "learning_rate": 9.200603318250378e-06,
-      "loss": 1.5325,
+      "grad_norm": 1.4854869842529297,
+      "learning_rate": 9.653092006033184e-06,
+      "loss": 1.5309,
       "step": 60
     },
     {
       "epoch": 0.5223880597014925,
-      "grad_norm": 2.2554094791412354,
-      "learning_rate": 9.049773755656109e-06,
-      "loss": 1.519,
+      "grad_norm": 2.791644811630249,
+      "learning_rate": 9.577677224736049e-06,
+      "loss": 1.4893,
       "step": 70
     },
     {
       "epoch": 0.5970149253731343,
-      "grad_norm": 3.507322072982788,
-      "learning_rate": 8.89894419306184e-06,
-      "loss": 1.3504,
+      "grad_norm": 3.8257875442504883,
+      "learning_rate": 9.502262443438914e-06,
+      "loss": 1.3155,
       "step": 80
     },
     {
       "epoch": 0.6716417910447762,
-      "grad_norm": 3.847669839859009,
-      "learning_rate": 8.748114630467572e-06,
-      "loss": 1.3068,
+      "grad_norm": 3.9002480506896973,
+      "learning_rate": 9.42684766214178e-06,
+      "loss": 1.3081,
       "step": 90
     },
     {
       "epoch": 0.746268656716418,
-      "grad_norm": 3.484496593475342,
-      "learning_rate": 8.597285067873304e-06,
-      "loss": 1.1601,
+      "grad_norm": 3.4716694355010986,
+      "learning_rate": 9.351432880844647e-06,
+      "loss": 1.1525,
       "step": 100
     },
     {
       "epoch": 0.746268656716418,
-      "eval_f1": 0.32132527894248475,
-      "eval_loss": 1.2031526565551758,
-      "eval_runtime": 2.6157,
-      "eval_samples_per_second": 420.916,
-      "eval_steps_per_second": 6.881,
+      "eval_f1": 0.33531687582947917,
+      "eval_loss": 1.234781265258789,
+      "eval_runtime": 2.6218,
+      "eval_samples_per_second": 419.936,
+      "eval_steps_per_second": 6.865,
       "step": 100
     },
     {
       "epoch": 0.8208955223880597,
-      "grad_norm": 3.5520474910736084,
-      "learning_rate": 8.446455505279036e-06,
-      "loss": 1.1665,
+      "grad_norm": 4.980569839477539,
+      "learning_rate": 9.276018099547513e-06,
+      "loss": 1.1661,
       "step": 110
     },
     {
       "epoch": 0.8955223880597015,
-      "grad_norm": 4.016570091247559,
-      "learning_rate": 8.295625942684767e-06,
-      "loss": 1.1374,
+      "grad_norm": 3.7840073108673096,
+      "learning_rate": 9.200603318250378e-06,
+      "loss": 1.1281,
       "step": 120
     },
     {
       "epoch": 0.9701492537313433,
-      "grad_norm": 4.674005508422852,
-      "learning_rate": 8.144796380090498e-06,
-      "loss": 1.1153,
+      "grad_norm": 4.9743733406066895,
+      "learning_rate": 9.125188536953243e-06,
+      "loss": 1.1082,
       "step": 130
     },
     {
       "epoch": 1.044776119402985,
-      "grad_norm": 4.79959774017334,
-      "learning_rate": 7.993966817496231e-06,
-      "loss": 1.1051,
+      "grad_norm": 4.535284042358398,
+      "learning_rate": 9.049773755656109e-06,
+      "loss": 1.0992,
       "step": 140
     },
     {
       "epoch": 1.1194029850746268,
-      "grad_norm": 4.5843000411987305,
-      "learning_rate": 7.84313725490196e-06,
-      "loss": 1.0437,
+      "grad_norm": 4.46350622177124,
+      "learning_rate": 8.974358974358976e-06,
+      "loss": 1.0405,
       "step": 150
     },
     {
       "epoch": 1.1940298507462686,
-      "grad_norm": 3.6620404720306396,
-      "learning_rate": 7.692307692307694e-06,
-      "loss": 1.0625,
+      "grad_norm": 3.532313823699951,
+      "learning_rate": 8.89894419306184e-06,
+      "loss": 1.0569,
       "step": 160
     },
     {
       "epoch": 1.2686567164179103,
-      "grad_norm": 4.64156436920166,
-      "learning_rate": 7.541478129713424e-06,
-      "loss": 0.9842,
+      "grad_norm": 4.936607360839844,
+      "learning_rate": 8.823529411764707e-06,
+      "loss": 0.981,
       "step": 170
     },
     {
       "epoch": 1.3432835820895521,
-      "grad_norm": 4.875860691070557,
-      "learning_rate": 7.390648567119156e-06,
-      "loss": 0.959,
+      "grad_norm": 4.794571876525879,
+      "learning_rate": 8.748114630467572e-06,
+      "loss": 0.9513,
       "step": 180
     },
     {
       "epoch": 1.417910447761194,
-      "grad_norm": 5.9996256828308105,
-      "learning_rate": 7.239819004524887e-06,
-      "loss": 1.0789,
+      "grad_norm": 5.570947170257568,
+      "learning_rate": 8.672699849170438e-06,
+      "loss": 1.0736,
       "step": 190
     },
     {
       "epoch": 1.4925373134328357,
-      "grad_norm": 6.207852363586426,
-      "learning_rate": 7.088989441930619e-06,
-      "loss": 1.0105,
+      "grad_norm": 8.54975700378418,
+      "learning_rate": 8.597285067873304e-06,
+      "loss": 1.0138,
       "step": 200
     },
     {
       "epoch": 1.4925373134328357,
-      "eval_f1": 0.49996652110080575,
-      "eval_loss": 1.0692095756530762,
-      "eval_runtime": 2.6262,
-      "eval_samples_per_second": 419.241,
-      "eval_steps_per_second": 6.854,
+      "eval_f1": 0.49237557928218106,
+      "eval_loss": 1.0882714986801147,
+      "eval_runtime": 2.6249,
+      "eval_samples_per_second": 419.441,
+      "eval_steps_per_second": 6.857,
       "step": 200
     },
     {
       "epoch": 1.5671641791044775,
-      "grad_norm": 4.4517292976379395,
-      "learning_rate": 6.938159879336351e-06,
-      "loss": 0.9994,
+      "grad_norm": 4.503359794616699,
+      "learning_rate": 8.52187028657617e-06,
+      "loss": 0.993,
       "step": 210
     },
     {
       "epoch": 1.6417910447761193,
-      "grad_norm": 4.6450395584106445,
-      "learning_rate": 6.787330316742083e-06,
-      "loss": 1.0023,
+      "grad_norm": 4.575403213500977,
+      "learning_rate": 8.446455505279036e-06,
+      "loss": 0.9965,
       "step": 220
     },
     {
       "epoch": 1.716417910447761,
-      "grad_norm": 5.14109992980957,
-      "learning_rate": 6.636500754147813e-06,
-      "loss": 1.0361,
+      "grad_norm": 6.2024736404418945,
+      "learning_rate": 8.371040723981902e-06,
+      "loss": 1.02,
       "step": 230
     },
     {
       "epoch": 1.7910447761194028,
-      "grad_norm": 4.552748203277588,
-      "learning_rate": 6.485671191553545e-06,
-      "loss": 1.0162,
+      "grad_norm": 4.787175178527832,
+      "learning_rate": 8.295625942684767e-06,
+      "loss": 1.013,
       "step": 240
     },
     {
       "epoch": 1.8656716417910446,
-      "grad_norm": 4.929900169372559,
-      "learning_rate": 6.334841628959276e-06,
-      "loss": 0.9702,
+      "grad_norm": 4.914880275726318,
+      "learning_rate": 8.220211161387632e-06,
+      "loss": 0.9593,
       "step": 250
     },
     {
       "epoch": 1.9402985074626866,
-      "grad_norm": 5.319007873535156,
-      "learning_rate": 6.1840120663650085e-06,
-      "loss": 1.0253,
+      "grad_norm": 5.331396579742432,
+      "learning_rate": 8.144796380090498e-06,
+      "loss": 1.0256,
       "step": 260
     },
     {
       "epoch": 2.014925373134328,
-      "grad_norm": 4.696887493133545,
-      "learning_rate": 6.033182503770739e-06,
-      "loss": 0.9516,
+      "grad_norm": 4.631103515625,
+      "learning_rate": 8.069381598793365e-06,
+      "loss": 0.9429,
       "step": 270
     },
     {
       "epoch": 2.08955223880597,
-      "grad_norm": 6.096177577972412,
-      "learning_rate": 5.882352941176471e-06,
-      "loss": 0.9563,
+      "grad_norm": 5.958868026733398,
+      "learning_rate": 7.993966817496231e-06,
+      "loss": 0.9433,
       "step": 280
     },
     {
       "epoch": 2.1641791044776117,
-      "grad_norm": 7.389795303344727,
-      "learning_rate": 5.731523378582202e-06,
-      "loss": 0.8915,
+      "grad_norm": 6.800838947296143,
+      "learning_rate": 7.918552036199096e-06,
+      "loss": 0.8713,
       "step": 290
     },
     {
       "epoch": 2.2388059701492535,
-      "grad_norm": 5.960943222045898,
-      "learning_rate": 5.580693815987934e-06,
-      "loss": 0.9256,
+      "grad_norm": 7.55411434173584,
+      "learning_rate": 7.84313725490196e-06,
+      "loss": 0.9136,
       "step": 300
     },
     {
       "epoch": 2.2388059701492535,
-      "eval_f1": 0.5184075253274552,
-      "eval_loss": 1.0643326044082642,
-      "eval_runtime": 2.6248,
-      "eval_samples_per_second": 419.458,
-      "eval_steps_per_second": 6.858,
+      "eval_f1": 0.5172149141468012,
+      "eval_loss": 1.072735071182251,
+      "eval_runtime": 2.6163,
+      "eval_samples_per_second": 420.822,
+      "eval_steps_per_second": 6.88,
       "step": 300
     },
     {
       "epoch": 2.3134328358208958,
-      "grad_norm": 5.415516376495361,
-      "learning_rate": 5.4298642533936655e-06,
-      "loss": 0.8906,
+      "grad_norm": 7.096745014190674,
+      "learning_rate": 7.767722473604827e-06,
+      "loss": 0.8746,
       "step": 310
     },
     {
       "epoch": 2.388059701492537,
-      "grad_norm": 6.378656387329102,
-      "learning_rate": 5.2790346907993975e-06,
-      "loss": 0.9267,
+      "grad_norm": 6.328619480133057,
+      "learning_rate": 7.692307692307694e-06,
+      "loss": 0.9159,
       "step": 320
     },
     {
       "epoch": 2.4626865671641793,
-      "grad_norm": 4.961767673492432,
-      "learning_rate": 5.128205128205128e-06,
-      "loss": 0.8908,
+      "grad_norm": 5.320040702819824,
+      "learning_rate": 7.616892911010558e-06,
+      "loss": 0.8736,
       "step": 330
     },
     {
       "epoch": 2.5373134328358207,
-      "grad_norm": 4.836255073547363,
-      "learning_rate": 4.97737556561086e-06,
-      "loss": 0.8994,
+      "grad_norm": 4.176707744598389,
+      "learning_rate": 7.541478129713424e-06,
+      "loss": 0.8954,
       "step": 340
     },
     {
       "epoch": 2.611940298507463,
-      "grad_norm": 6.633901596069336,
-      "learning_rate": 4.826546003016592e-06,
-      "loss": 0.965,
+      "grad_norm": 6.141348838806152,
+      "learning_rate": 7.4660633484162904e-06,
+      "loss": 0.9559,
       "step": 350
     },
     {
       "epoch": 2.6865671641791042,
-      "grad_norm": 6.069859504699707,
-      "learning_rate": 4.675716440422323e-06,
-      "loss": 0.8993,
+      "grad_norm": 5.281525135040283,
+      "learning_rate": 7.390648567119156e-06,
+      "loss": 0.8859,
       "step": 360
     },
     {
       "epoch": 2.7611940298507465,
-      "grad_norm": 7.4525065422058105,
-      "learning_rate": 4.5248868778280546e-06,
-      "loss": 0.8591,
+      "grad_norm": 6.908074378967285,
+      "learning_rate": 7.315233785822022e-06,
+      "loss": 0.8344,
       "step": 370
     },
     {
       "epoch": 2.835820895522388,
-      "grad_norm": 7.40543794631958,
-      "learning_rate": 4.374057315233786e-06,
-      "loss": 0.9181,
+      "grad_norm": 7.840199947357178,
+      "learning_rate": 7.239819004524887e-06,
+      "loss": 0.8987,
       "step": 380
     },
     {
       "epoch": 2.91044776119403,
-      "grad_norm": 4.696629047393799,
-      "learning_rate": 4.223227752639518e-06,
-      "loss": 0.9814,
+      "grad_norm": 5.066633224487305,
+      "learning_rate": 7.164404223227753e-06,
+      "loss": 0.9643,
       "step": 390
     },
     {
       "epoch": 2.9850746268656714,
-      "grad_norm": 5.66481351852417,
-      "learning_rate": 4.072398190045249e-06,
-      "loss": 0.8987,
+      "grad_norm": 7.574981689453125,
+      "learning_rate": 7.088989441930619e-06,
+      "loss": 0.8806,
       "step": 400
     },
     {
       "epoch": 2.9850746268656714,
-      "eval_f1": 0.5115649905281687,
-      "eval_loss": 1.1100599765777588,
-      "eval_runtime": 2.6322,
-      "eval_samples_per_second": 418.288,
-      "eval_steps_per_second": 6.838,
+      "eval_f1": 0.5204655099703751,
+      "eval_loss": 1.100539207458496,
+      "eval_runtime": 2.6156,
+      "eval_samples_per_second": 420.93,
+      "eval_steps_per_second": 6.882,
       "step": 400
     },
     {
       "epoch": 3.0597014925373136,
-      "grad_norm": 6.0242719650268555,
-      "learning_rate": 3.92156862745098e-06,
-      "loss": 0.8692,
+      "grad_norm": 5.5668864250183105,
+      "learning_rate": 7.013574660633484e-06,
+      "loss": 0.8348,
       "step": 410
     },
     {
       "epoch": 3.1343283582089554,
-      "grad_norm": 5.19691276550293,
-      "learning_rate": 3.770739064856712e-06,
-      "loss": 0.8274,
+      "grad_norm": 6.508656024932861,
+      "learning_rate": 6.938159879336351e-06,
+      "loss": 0.7917,
       "step": 420
     },
     {
       "epoch": 3.208955223880597,
-      "grad_norm": 4.793543815612793,
-      "learning_rate": 3.6199095022624436e-06,
-      "loss": 0.8599,
+      "grad_norm": 4.860711097717285,
+      "learning_rate": 6.862745098039216e-06,
+      "loss": 0.8259,
       "step": 430
     },
     {
       "epoch": 3.283582089552239,
-      "grad_norm": 6.505572319030762,
-      "learning_rate": 3.4690799396681753e-06,
-      "loss": 0.8684,
+      "grad_norm": 6.06543493270874,
+      "learning_rate": 6.787330316742083e-06,
+      "loss": 0.8361,
       "step": 440
     },
     {
       "epoch": 3.3582089552238807,
-      "grad_norm": 7.902689456939697,
-      "learning_rate": 3.3182503770739065e-06,
-      "loss": 0.8503,
+      "grad_norm": 7.322782516479492,
+      "learning_rate": 6.7119155354449474e-06,
+      "loss": 0.8154,
       "step": 450
     },
     {
       "epoch": 3.4328358208955225,
-      "grad_norm": 7.128357410430908,
-      "learning_rate": 3.167420814479638e-06,
-      "loss": 0.827,
+      "grad_norm": 7.006171703338623,
+      "learning_rate": 6.636500754147813e-06,
+      "loss": 0.796,
       "step": 460
     },
     {
       "epoch": 3.5074626865671643,
-      "grad_norm": 4.8343400955200195,
-      "learning_rate": 3.0165912518853694e-06,
-      "loss": 0.8132,
+      "grad_norm": 5.344196796417236,
+      "learning_rate": 6.5610859728506795e-06,
+      "loss": 0.7981,
       "step": 470
     },
     {
       "epoch": 3.582089552238806,
-      "grad_norm": 4.999044895172119,
-      "learning_rate": 2.865761689291101e-06,
-      "loss": 0.8746,
+      "grad_norm": 6.2761735916137695,
+      "learning_rate": 6.485671191553545e-06,
+      "loss": 0.8443,
       "step": 480
     },
     {
       "epoch": 3.656716417910448,
-      "grad_norm": 6.041540145874023,
-      "learning_rate": 2.7149321266968327e-06,
-      "loss": 0.8541,
+      "grad_norm": 5.846128463745117,
+      "learning_rate": 6.410256410256412e-06,
+      "loss": 0.8236,
       "step": 490
     },
     {
       "epoch": 3.7313432835820897,
-      "grad_norm": 5.733530521392822,
-      "learning_rate": 2.564102564102564e-06,
-      "loss": 0.8379,
+      "grad_norm": 6.686851501464844,
+      "learning_rate": 6.334841628959276e-06,
+      "loss": 0.8115,
       "step": 500
     },
     {
       "epoch": 3.7313432835820897,
-      "eval_f1": 0.5254742032963681,
-      "eval_loss": 1.068690538406372,
-      "eval_runtime": 2.6268,
-      "eval_samples_per_second": 419.14,
-      "eval_steps_per_second": 6.852,
+      "eval_f1": 0.5492932407699946,
+      "eval_loss": 1.054356575012207,
+      "eval_runtime": 2.6244,
+      "eval_samples_per_second": 419.531,
+      "eval_steps_per_second": 6.859,
       "step": 500
     },
     {
       "epoch": 3.8059701492537314,
-      "grad_norm": 5.6245951652526855,
-      "learning_rate": 2.413273001508296e-06,
-      "loss": 0.8214,
+      "grad_norm": 6.415255069732666,
+      "learning_rate": 6.259426847662142e-06,
+      "loss": 0.7944,
       "step": 510
     },
     {
       "epoch": 3.8805970149253732,
-      "grad_norm": 8.329957008361816,
-      "learning_rate": 2.2624434389140273e-06,
-      "loss": 0.8727,
+      "grad_norm": 8.690467834472656,
+      "learning_rate": 6.1840120663650085e-06,
+      "loss": 0.8418,
       "step": 520
     },
     {
       "epoch": 3.955223880597015,
-      "grad_norm": 6.278369903564453,
-      "learning_rate": 2.111613876319759e-06,
-      "loss": 0.8601,
+      "grad_norm": 16.526342391967773,
+      "learning_rate": 6.108597285067874e-06,
+      "loss": 0.8438,
       "step": 530
     },
     {
       "epoch": 4.029850746268656,
-      "grad_norm": 4.968955039978027,
-      "learning_rate": 1.96078431372549e-06,
-      "loss": 0.8378,
+      "grad_norm": 5.72576904296875,
+      "learning_rate": 6.033182503770739e-06,
+      "loss": 0.7861,
       "step": 540
     },
     {
       "epoch": 4.104477611940299,
-      "grad_norm": 7.428950786590576,
-      "learning_rate": 1.8099547511312218e-06,
-      "loss": 0.8036,
+      "grad_norm": 6.920104026794434,
+      "learning_rate": 5.957767722473605e-06,
+      "loss": 0.7423,
       "step": 550
     },
     {
       "epoch": 4.17910447761194,
-      "grad_norm": 4.700170516967773,
-      "learning_rate": 1.6591251885369533e-06,
-      "loss": 0.8061,
+      "grad_norm": 4.901302337646484,
+      "learning_rate": 5.882352941176471e-06,
+      "loss": 0.7375,
       "step": 560
     },
     {
       "epoch": 4.253731343283582,
-      "grad_norm": 6.555200576782227,
-      "learning_rate": 1.5082956259426847e-06,
-      "loss": 0.8678,
+      "grad_norm": 9.393866539001465,
+      "learning_rate": 5.806938159879337e-06,
+      "loss": 0.7803,
       "step": 570
     },
     {
       "epoch": 4.3283582089552235,
-      "grad_norm": 5.239262104034424,
-      "learning_rate": 1.3574660633484164e-06,
-      "loss": 0.8529,
+      "grad_norm": 6.509814739227295,
+      "learning_rate": 5.731523378582202e-06,
+      "loss": 0.8197,
       "step": 580
     },
     {
       "epoch": 4.402985074626866,
-      "grad_norm": 5.288607120513916,
-      "learning_rate": 1.206636500754148e-06,
-      "loss": 0.8099,
+      "grad_norm": 7.71168851852417,
+      "learning_rate": 5.656108597285069e-06,
+      "loss": 0.7656,
       "step": 590
     },
     {
       "epoch": 4.477611940298507,
-      "grad_norm": 5.942199230194092,
-      "learning_rate": 1.0558069381598795e-06,
-      "loss": 0.8048,
+      "grad_norm": 8.574570655822754,
+      "learning_rate": 5.580693815987934e-06,
+      "loss": 0.7483,
       "step": 600
     },
     {
       "epoch": 4.477611940298507,
-      "eval_f1": 0.531250378962142,
-      "eval_loss": 1.0681172609329224,
-      "eval_runtime": 2.6271,
-      "eval_samples_per_second": 419.101,
-      "eval_steps_per_second": 6.852,
+      "eval_f1": 0.5307248014454118,
+      "eval_loss": 1.1092573404312134,
+      "eval_runtime": 2.6211,
+      "eval_samples_per_second": 420.049,
+      "eval_steps_per_second": 6.867,
       "step": 600
     },
     {
       "epoch": 4.552238805970149,
-      "grad_norm": 6.185580253601074,
-      "learning_rate": 9.049773755656109e-07,
-      "loss": 0.8314,
+      "grad_norm": 6.70869779586792,
+      "learning_rate": 5.505279034690801e-06,
+      "loss": 0.7771,
       "step": 610
     },
     {
       "epoch": 4.6268656716417915,
-      "grad_norm": 6.450467109680176,
-      "learning_rate": 7.541478129713424e-07,
-      "loss": 0.7659,
+      "grad_norm": 7.6885480880737305,
+      "learning_rate": 5.4298642533936655e-06,
+      "loss": 0.7165,
       "step": 620
     },
     {
       "epoch": 4.701492537313433,
-      "grad_norm": 4.966616630554199,
-      "learning_rate": 6.03318250377074e-07,
-      "loss": 0.7824,
+      "grad_norm": 6.220367908477783,
+      "learning_rate": 5.354449472096531e-06,
+      "loss": 0.7245,
       "step": 630
     },
     {
       "epoch": 4.776119402985074,
-      "grad_norm": 5.365437030792236,
-      "learning_rate": 4.5248868778280546e-07,
-      "loss": 0.7985,
+      "grad_norm": 7.706441879272461,
+      "learning_rate": 5.2790346907993975e-06,
+      "loss": 0.7412,
       "step": 640
     },
     {
       "epoch": 4.850746268656716,
-      "grad_norm": 5.803382396697998,
-      "learning_rate": 3.01659125188537e-07,
-      "loss": 0.7774,
+      "grad_norm": 6.295016765594482,
+      "learning_rate": 5.203619909502263e-06,
+      "loss": 0.723,
       "step": 650
     },
     {
       "epoch": 4.925373134328359,
-      "grad_norm": 5.00211763381958,
-      "learning_rate": 1.508295625942685e-07,
-      "loss": 0.8225,
+      "grad_norm": 5.999417781829834,
+      "learning_rate": 5.128205128205128e-06,
+      "loss": 0.7665,
       "step": 660
     },
     {
       "epoch": 5.0,
-      "grad_norm": 11.036751747131348,
-      "learning_rate": 0.0,
-      "loss": 0.8557,
+      "grad_norm": 11.957300186157227,
+      "learning_rate": 5.052790346907994e-06,
+      "loss": 0.7852,
       "step": 670
     },
     {
-      "epoch": 5.0,
-      "step": 670,
-      "total_flos": 6586294115808000.0,
-      "train_loss": 0.9968819860202163,
-      "train_runtime": 296.7277,
-      "train_samples_per_second": 143.97,
-      "train_steps_per_second": 2.258
+      "epoch": 5.074626865671641,
+      "grad_norm": 8.489127159118652,
+      "learning_rate": 4.97737556561086e-06,
+      "loss": 0.7029,
+      "step": 680
+    },
+    {
+      "epoch": 5.149253731343284,
+      "grad_norm": 9.558174133300781,
+      "learning_rate": 4.901960784313726e-06,
+      "loss": 0.7006,
+      "step": 690
+    },
+    {
+      "epoch": 5.223880597014926,
+      "grad_norm": 5.191695213317871,
+      "learning_rate": 4.826546003016592e-06,
+      "loss": 0.696,
+      "step": 700
+    },
+    {
+      "epoch": 5.223880597014926,
+      "eval_f1": 0.5424372891596922,
+      "eval_loss": 1.1064388751983643,
+      "eval_runtime": 2.626,
+      "eval_samples_per_second": 419.262,
+      "eval_steps_per_second": 6.854,
+      "step": 700
+    },
+    {
+      "epoch": 5.298507462686567,
+      "grad_norm": 10.742992401123047,
+      "learning_rate": 4.751131221719457e-06,
+      "loss": 0.7083,
+      "step": 710
+    },
+    {
+      "epoch": 5.373134328358209,
+      "grad_norm": 7.179316997528076,
+      "learning_rate": 4.675716440422323e-06,
+      "loss": 0.6909,
+      "step": 720
+    },
+    {
+      "epoch": 5.447761194029851,
+      "grad_norm": 8.081168174743652,
+      "learning_rate": 4.600301659125189e-06,
+      "loss": 0.7018,
+      "step": 730
+    },
+    {
+      "epoch": 5.522388059701493,
+      "grad_norm": 7.361329555511475,
+      "learning_rate": 4.5248868778280546e-06,
+      "loss": 0.7003,
+      "step": 740
+    },
+    {
+      "epoch": 5.597014925373134,
+      "grad_norm": 12.964821815490723,
+      "learning_rate": 4.44947209653092e-06,
+      "loss": 0.716,
+      "step": 750
+    },
+    {
+      "epoch": 5.6716417910447765,
+      "grad_norm": 7.4841084480285645,
+      "learning_rate": 4.374057315233786e-06,
+      "loss": 0.7114,
+      "step": 760
+    },
+    {
+      "epoch": 5.746268656716418,
+      "grad_norm": 7.906242370605469,
+      "learning_rate": 4.298642533936652e-06,
+      "loss": 0.7097,
+      "step": 770
+    },
+    {
+      "epoch": 5.82089552238806,
+      "grad_norm": 8.054797172546387,
+      "learning_rate": 4.223227752639518e-06,
+      "loss": 0.6926,
+      "step": 780
+    },
+    {
+      "epoch": 5.895522388059701,
+      "grad_norm": 6.4073805809021,
+      "learning_rate": 4.1478129713423835e-06,
+      "loss": 0.7081,
+      "step": 790
+    },
+    {
+      "epoch": 5.970149253731344,
+      "grad_norm": 7.464499473571777,
+      "learning_rate": 4.072398190045249e-06,
+      "loss": 0.7167,
+      "step": 800
+    },
+    {
+      "epoch": 5.970149253731344,
+      "eval_f1": 0.5292750734436421,
+      "eval_loss": 1.1407464742660522,
+      "eval_runtime": 2.6282,
+      "eval_samples_per_second": 418.913,
+      "eval_steps_per_second": 6.849,
+      "step": 800
+    },
+    {
+      "epoch": 6.044776119402985,
+      "grad_norm": 7.362347602844238,
+      "learning_rate": 3.9969834087481156e-06,
+      "loss": 0.6765,
+      "step": 810
+    },
+    {
+      "epoch": 6.119402985074627,
+      "grad_norm": 6.4378743171691895,
+      "learning_rate": 3.92156862745098e-06,
+      "loss": 0.667,
+      "step": 820
+    },
+    {
+      "epoch": 6.1940298507462686,
+      "grad_norm": 6.983713150024414,
+      "learning_rate": 3.846153846153847e-06,
+      "loss": 0.6804,
+      "step": 830
+    },
+    {
+      "epoch": 6.268656716417911,
+      "grad_norm": 6.926436901092529,
+      "learning_rate": 3.770739064856712e-06,
+      "loss": 0.6945,
+      "step": 840
+    },
+    {
+      "epoch": 6.343283582089552,
+      "grad_norm": 7.79516077041626,
+      "learning_rate": 3.695324283559578e-06,
+      "loss": 0.6491,
+      "step": 850
+    },
+    {
+      "epoch": 6.417910447761194,
+      "grad_norm": 6.210599899291992,
+      "learning_rate": 3.6199095022624436e-06,
+      "loss": 0.6564,
+      "step": 860
+    },
+    {
+      "epoch": 6.492537313432836,
+      "grad_norm": 10.464361190795898,
+      "learning_rate": 3.5444947209653097e-06,
+      "loss": 0.6735,
+      "step": 870
+    },
+    {
+      "epoch": 6.567164179104478,
+      "grad_norm": 8.545991897583008,
+      "learning_rate": 3.4690799396681753e-06,
+      "loss": 0.6456,
+      "step": 880
+    },
+    {
+      "epoch": 6.641791044776119,
+      "grad_norm": 7.849530220031738,
+      "learning_rate": 3.3936651583710413e-06,
+      "loss": 0.6301,
+      "step": 890
+    },
+    {
+      "epoch": 6.7164179104477615,
+      "grad_norm": 8.215511322021484,
+      "learning_rate": 3.3182503770739065e-06,
+      "loss": 0.7079,
+      "step": 900
+    },
+    {
+      "epoch": 6.7164179104477615,
+      "eval_f1": 0.5206113722605857,
+      "eval_loss": 1.1766188144683838,
+      "eval_runtime": 2.6268,
+      "eval_samples_per_second": 419.139,
+      "eval_steps_per_second": 6.852,
+      "step": 900
+    },
+    {
+      "epoch": 6.791044776119403,
+      "grad_norm": 5.998091220855713,
+      "learning_rate": 3.2428355957767726e-06,
+      "loss": 0.6949,
+      "step": 910
+    },
+    {
+      "epoch": 6.865671641791045,
+      "grad_norm": 7.981301784515381,
+      "learning_rate": 3.167420814479638e-06,
+      "loss": 0.6837,
+      "step": 920
+    },
+    {
+      "epoch": 6.940298507462686,
+      "grad_norm": 7.27372407913208,
+      "learning_rate": 3.0920060331825042e-06,
+      "loss": 0.6673,
+      "step": 930
+    },
+    {
+      "epoch": 7.014925373134329,
+      "grad_norm": 6.519421100616455,
+      "learning_rate": 3.0165912518853694e-06,
+      "loss": 0.6366,
+      "step": 940
+    },
+    {
+      "epoch": 7.08955223880597,
+      "grad_norm": 9.983951568603516,
+      "learning_rate": 2.9411764705882355e-06,
+      "loss": 0.6015,
+      "step": 950
+    },
+    {
+      "epoch": 7.164179104477612,
+      "grad_norm": 8.719895362854004,
+      "learning_rate": 2.865761689291101e-06,
+      "loss": 0.6228,
+      "step": 960
+    },
+    {
+      "epoch": 7.2388059701492535,
+      "grad_norm": 9.368388175964355,
+      "learning_rate": 2.790346907993967e-06,
+      "loss": 0.6213,
+      "step": 970
+    },
+    {
+      "epoch": 7.313432835820896,
+      "grad_norm": 6.723618984222412,
+      "learning_rate": 2.7149321266968327e-06,
+      "loss": 0.6307,
+      "step": 980
+    },
+    {
+      "epoch": 7.388059701492537,
+      "grad_norm": 6.052383899688721,
+      "learning_rate": 2.6395173453996988e-06,
+      "loss": 0.5769,
+      "step": 990
+    },
+    {
+      "epoch": 7.462686567164179,
+      "grad_norm": 6.24008846282959,
+      "learning_rate": 2.564102564102564e-06,
+      "loss": 0.6593,
+      "step": 1000
+    },
+    {
+      "epoch": 7.462686567164179,
+      "eval_f1": 0.5257267484677619,
+      "eval_loss": 1.2102895975112915,
+      "eval_runtime": 2.6248,
+      "eval_samples_per_second": 419.454,
+      "eval_steps_per_second": 6.858,
+      "step": 1000
+    },
+    {
+      "epoch": 7.537313432835821,
+      "grad_norm": 12.369339942932129,
+      "learning_rate": 2.48868778280543e-06,
+      "loss": 0.6706,
+      "step": 1010
+    },
+    {
+      "epoch": 7.611940298507463,
+      "grad_norm": 5.776430606842041,
+      "learning_rate": 2.413273001508296e-06,
+      "loss": 0.6241,
+      "step": 1020
+    },
+    {
+      "epoch": 7.686567164179104,
+      "grad_norm": 7.558428764343262,
+      "learning_rate": 2.3378582202111617e-06,
+      "loss": 0.6643,
+      "step": 1030
+    },
+    {
+      "epoch": 7.7611940298507465,
+      "grad_norm": 9.2826509475708,
+      "learning_rate": 2.2624434389140273e-06,
+      "loss": 0.6296,
+      "step": 1040
+    },
+    {
+      "epoch": 7.835820895522388,
+      "grad_norm": 7.237534999847412,
+      "learning_rate": 2.187028657616893e-06,
+      "loss": 0.6162,
+      "step": 1050
+    },
+    {
+      "epoch": 7.91044776119403,
+      "grad_norm": 6.077177047729492,
+      "learning_rate": 2.111613876319759e-06,
+      "loss": 0.6446,
+      "step": 1060
+    },
+    {
+      "epoch": 7.985074626865671,
+      "grad_norm": 7.066919803619385,
+      "learning_rate": 2.0361990950226245e-06,
+      "loss": 0.6244,
+      "step": 1070
+    },
+    {
+      "epoch": 8.059701492537313,
+      "grad_norm": 6.757171630859375,
+      "learning_rate": 1.96078431372549e-06,
+      "loss": 0.6031,
+      "step": 1080
+    },
+    {
+      "epoch": 8.134328358208956,
+      "grad_norm": 7.3823018074035645,
+      "learning_rate": 1.885369532428356e-06,
+      "loss": 0.5991,
+      "step": 1090
+    },
+    {
+      "epoch": 8.208955223880597,
+      "grad_norm": 7.998798370361328,
+      "learning_rate": 1.8099547511312218e-06,
+      "loss": 0.6299,
+      "step": 1100
+    },
+    {
+      "epoch": 8.208955223880597,
+      "eval_f1": 0.5248808380426293,
+      "eval_loss": 1.2031023502349854,
+      "eval_runtime": 2.624,
+      "eval_samples_per_second": 419.588,
+      "eval_steps_per_second": 6.86,
+      "step": 1100
+    },
+    {
+      "epoch": 8.283582089552239,
+      "grad_norm": 8.981494903564453,
+      "learning_rate": 1.7345399698340876e-06,
+      "loss": 0.6001,
+      "step": 1110
+    },
+    {
+      "epoch": 8.35820895522388,
+      "grad_norm": 8.06408977508545,
+      "learning_rate": 1.6591251885369533e-06,
+      "loss": 0.6103,
+      "step": 1120
+    },
+    {
+      "epoch": 8.432835820895523,
+      "grad_norm": 8.383085250854492,
+      "learning_rate": 1.583710407239819e-06,
+      "loss": 0.6015,
+      "step": 1130
+    },
+    {
+      "epoch": 8.507462686567164,
+      "grad_norm": 8.711113929748535,
+      "learning_rate": 1.5082956259426847e-06,
+      "loss": 0.5683,
+      "step": 1140
+    },
+    {
+      "epoch": 8.582089552238806,
+      "grad_norm": 6.439323902130127,
+      "learning_rate": 1.4328808446455505e-06,
+      "loss": 0.5873,
+      "step": 1150
+    },
+    {
+      "epoch": 8.656716417910447,
+      "grad_norm": 7.051445484161377,
+      "learning_rate": 1.3574660633484164e-06,
+      "loss": 0.6062,
+      "step": 1160
+    },
+    {
+      "epoch": 8.73134328358209,
+      "grad_norm": 10.478583335876465,
+      "learning_rate": 1.282051282051282e-06,
+      "loss": 0.5616,
+      "step": 1170
+    },
+    {
+      "epoch": 8.805970149253731,
+      "grad_norm": 8.301910400390625,
+      "learning_rate": 1.206636500754148e-06,
+      "loss": 0.579,
+      "step": 1180
+    },
+    {
+      "epoch": 8.880597014925373,
+      "grad_norm": 6.833534240722656,
+      "learning_rate": 1.1312217194570136e-06,
+      "loss": 0.6271,
+      "step": 1190
+    },
+    {
+      "epoch": 8.955223880597014,
+      "grad_norm": 12.211673736572266,
+      "learning_rate": 1.0558069381598795e-06,
+      "loss": 0.675,
+      "step": 1200
+    },
+    {
+      "epoch": 8.955223880597014,
+      "eval_f1": 0.5198122319190461,
+      "eval_loss": 1.2117300033569336,
+      "eval_runtime": 2.6239,
+      "eval_samples_per_second": 419.601,
+      "eval_steps_per_second": 6.86,
+      "step": 1200
+    },
+    {
+      "epoch": 9.029850746268657,
+      "grad_norm": 7.296724796295166,
+      "learning_rate": 9.80392156862745e-07,
+      "loss": 0.6175,
+      "step": 1210
+    },
+    {
+      "epoch": 9.104477611940299,
+      "grad_norm": 8.846306800842285,
+      "learning_rate": 9.049773755656109e-07,
+      "loss": 0.5649,
+      "step": 1220
+    },
+    {
+      "epoch": 9.17910447761194,
+      "grad_norm": 10.289523124694824,
+      "learning_rate": 8.295625942684766e-07,
+      "loss": 0.6298,
+      "step": 1230
+    },
+    {
+      "epoch": 9.253731343283581,
+      "grad_norm": 7.686677932739258,
+      "learning_rate": 7.541478129713424e-07,
+      "loss": 0.5787,
+      "step": 1240
+    },
+    {
+      "epoch": 9.328358208955224,
+      "grad_norm": 10.47728443145752,
+      "learning_rate": 6.787330316742082e-07,
+      "loss": 0.5792,
+      "step": 1250
+    },
+    {
+      "epoch": 9.402985074626866,
+      "grad_norm": 7.4656500816345215,
+      "learning_rate": 6.03318250377074e-07,
+      "loss": 0.6518,
+      "step": 1260
+    },
+    {
+      "epoch": 9.477611940298507,
+      "grad_norm": 6.7001051902771,
+      "learning_rate": 5.279034690799397e-07,
+      "loss": 0.5787,
+      "step": 1270
+    },
+    {
+      "epoch": 9.552238805970148,
+      "grad_norm": 8.94778823852539,
+      "learning_rate": 4.5248868778280546e-07,
+      "loss": 0.5844,
+      "step": 1280
+    },
+    {
+      "epoch": 9.626865671641792,
+      "grad_norm": 6.784146308898926,
+      "learning_rate": 3.770739064856712e-07,
+      "loss": 0.5857,
+      "step": 1290
+    },
+    {
+      "epoch": 9.701492537313433,
+      "grad_norm": 7.766072750091553,
+      "learning_rate": 3.01659125188537e-07,
+      "loss": 0.5619,
+      "step": 1300
+    },
+    {
+      "epoch": 9.701492537313433,
+      "eval_f1": 0.5208328976255492,
+      "eval_loss": 1.2219291925430298,
+      "eval_runtime": 2.6161,
+      "eval_samples_per_second": 420.863,
+      "eval_steps_per_second": 6.881,
+      "step": 1300
+    },
+    {
+      "epoch": 9.776119402985074,
+      "grad_norm": 7.4001874923706055,
+      "learning_rate": 2.2624434389140273e-07,
+      "loss": 0.5874,
+      "step": 1310
+    },
+    {
+      "epoch": 9.850746268656717,
+      "grad_norm": 7.507802486419678,
+      "learning_rate": 1.508295625942685e-07,
+      "loss": 0.5875,
+      "step": 1320
+    },
+    {
+      "epoch": 9.925373134328359,
+      "grad_norm": 6.6053147315979,
+      "learning_rate": 7.541478129713425e-08,
+      "loss": 0.5564,
+      "step": 1330
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 8.790937423706055,
+      "learning_rate": 0.0,
+      "loss": 0.5623,
+      "step": 1340
+    },
+    {
+      "epoch": 10.0,
+      "step": 1340,
+      "total_flos": 1.3172588231616e+16,
+      "train_loss": 0.8062449658094947,
+      "train_runtime": 588.6173,
+      "train_samples_per_second": 145.154,
+      "train_steps_per_second": 2.277
     }
   ],
   "logging_steps": 10,
-  "max_steps": 670,
+  "max_steps": 1340,
   "num_input_tokens_seen": 0,
-  "num_train_epochs": 5,
+  "num_train_epochs": 10,
   "save_steps": 100,
   "stateful_callbacks": {
     "TrainerControl": {
@@ -558,7 +1090,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 6586294115808000.0,
+  "total_flos": 1.3172588231616e+16,
   "train_batch_size": 64,
   "trial_name": null,
   "trial_params": null