Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -1073,3 +1073,8 @@
 {"step": 21460, "tokens": 11251220480, "loss": 2.7553260803222654, "tokens_per_second": 942099.8717380728, "lr": 0.0015, "gradient_norm": 0.17784038186073303, "benchmark": false}
 {"step": 21480, "tokens": 11261706240, "loss": 2.721171569824219, "tokens_per_second": 984823.765624844, "lr": 0.0015, "gradient_norm": 0.1572302281856537, "benchmark": false}
 {"step": 21500, "tokens": 11272192000, "loss": 2.7307083129882814, "tokens_per_second": 977356.4326711252, "lr": 0.0015, "gradient_norm": 0.1457183063030243, "benchmark": false}
+{"step": 21520, "tokens": 11282677760, "loss": 2.739332580566406, "tokens_per_second": 861968.9537989718, "lr": 0.0015, "gradient_norm": 0.1666897088289261, "benchmark": false}
+{"step": 21540, "tokens": 11293163520, "loss": 2.6986692428588865, "tokens_per_second": 989445.1097467432, "lr": 0.0015, "gradient_norm": 0.1420649290084839, "benchmark": false}
+{"step": 21560, "tokens": 11303649280, "loss": 2.75574951171875, "tokens_per_second": 988670.3337918094, "lr": 0.0015, "gradient_norm": 0.1459985226392746, "benchmark": false}
+{"step": 21580, "tokens": 11314135040, "loss": 2.728601837158203, "tokens_per_second": 814876.4121760112, "lr": 0.0015, "gradient_norm": 0.16379660367965698, "benchmark": false}
+{"step": 21600, "tokens": 11324620800, "loss": 2.744864654541016, "tokens_per_second": 982885.1191783481, "lr": 0.0015, "gradient_norm": 0.17417514324188232, "benchmark": false}