Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -48,3 +48,7 @@
 {"step": 960, "tokens": 503316480, "loss": 3.5565567016601562, "tokens_per_second": 991150.2588607525, "lr": 0.0007199999999999999, "gradient_norm": 0.3613435924053192, "benchmark": false}
 {"step": 980, "tokens": 513802240, "loss": 3.534979248046875, "tokens_per_second": 954816.5784302262, "lr": 0.000735, "gradient_norm": 0.36277028918266296, "benchmark": false}
 {"step": 1000, "tokens": 524288000, "loss": 3.512640380859375, "tokens_per_second": 982882.901952814, "lr": 0.00075, "gradient_norm": 0.36223894357681274, "benchmark": false}
+{"step": 1020, "tokens": 534773760, "loss": 3.498727798461914, "tokens_per_second": 788526.1090729708, "lr": 0.0007650000000000001, "gradient_norm": 0.33503326773643494, "benchmark": false}
+{"step": 1040, "tokens": 545259520, "loss": 3.485460662841797, "tokens_per_second": 996745.0972588192, "lr": 0.0007800000000000001, "gradient_norm": 0.3457220494747162, "benchmark": false}
+{"step": 1060, "tokens": 555745280, "loss": 3.48607177734375, "tokens_per_second": 996269.3046775112, "lr": 0.000795, "gradient_norm": 0.37579262256622314, "benchmark": false}
+{"step": 1080, "tokens": 566231040, "loss": 3.5009490966796877, "tokens_per_second": 993456.0662475126, "lr": 0.0008100000000000001, "gradient_norm": 0.3382250666618347, "benchmark": false}