Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -373,3 +373,7 @@
 {"step": 7460, "tokens": 3911188480, "loss": 2.868077850341797, "tokens_per_second": 971001.3676481111, "lr": 0.0015, "gradient_norm": 0.0777946189045906, "benchmark": false}
 {"step": 7480, "tokens": 3921674240, "loss": 2.8896739959716795, "tokens_per_second": 971052.3362966019, "lr": 0.0015, "gradient_norm": 0.06711418926715851, "benchmark": false}
 {"step": 7500, "tokens": 3932160000, "loss": 2.8810169219970705, "tokens_per_second": 833273.9147864264, "lr": 0.0015, "gradient_norm": 0.06997272372245789, "benchmark": false}
+{"step": 7520, "tokens": 3942645760, "loss": 2.8911417007446287, "tokens_per_second": 858002.5052553657, "lr": 0.0015, "gradient_norm": 0.07078112661838531, "benchmark": false}
+{"step": 7540, "tokens": 3953131520, "loss": 2.867593765258789, "tokens_per_second": 992329.0947803116, "lr": 0.0015, "gradient_norm": 0.07130780071020126, "benchmark": false}
+{"step": 7560, "tokens": 3963617280, "loss": 2.8703800201416017, "tokens_per_second": 981001.7008508949, "lr": 0.0015, "gradient_norm": 0.07841914147138596, "benchmark": false}
+{"step": 7580, "tokens": 3974103040, "loss": 2.8924484252929688, "tokens_per_second": 983234.1127777771, "lr": 0.0015, "gradient_norm": 0.06483935564756393, "benchmark": false}