Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -148,3 +148,7 @@
 {"step": 2960, "tokens": 1551892480, "loss": 3.0917444229125977, "tokens_per_second": 976263.4119847118, "lr": 0.0015, "gradient_norm": 0.16529427468776703, "benchmark": false}
 {"step": 2980, "tokens": 1562378240, "loss": 3.107547378540039, "tokens_per_second": 974664.1426940245, "lr": 0.0015, "gradient_norm": 0.1620856076478958, "benchmark": false}
 {"step": 3000, "tokens": 1572864000, "loss": 3.053942108154297, "tokens_per_second": 961120.6402985074, "lr": 0.0015, "gradient_norm": 0.17983736097812653, "benchmark": false}
+{"step": 3020, "tokens": 1583349760, "loss": 3.0546281814575194, "tokens_per_second": 812529.1170383674, "lr": 0.0015, "gradient_norm": 0.13785792887210846, "benchmark": false}
+{"step": 3040, "tokens": 1593835520, "loss": 3.0774038314819334, "tokens_per_second": 995183.5208714989, "lr": 0.0015, "gradient_norm": 0.18102078139781952, "benchmark": false}
+{"step": 3060, "tokens": 1604321280, "loss": 3.0798330307006836, "tokens_per_second": 804488.664425958, "lr": 0.0015, "gradient_norm": 0.1693069338798523, "benchmark": false}
+{"step": 3080, "tokens": 1614807040, "loss": 3.0429784774780275, "tokens_per_second": 979478.5009066728, "lr": 0.0015, "gradient_norm": 0.1676941066980362, "benchmark": false}