Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -624,3 +624,8 @@
 {"step": 12480, "tokens": 6543114240, "loss": 2.778891944885254, "tokens_per_second": 850421.2073633118, "lr": 0.0015, "gradient_norm": 0.11233720183372498, "benchmark": false}
 {"step": 12500, "tokens": 6553600000, "loss": 2.7939947128295897, "tokens_per_second": 973450.8494435991, "lr": 0.0015, "gradient_norm": 0.1016482263803482, "benchmark": false}
 {"step": 12520, "tokens": 6564085760, "loss": 2.7848548889160156, "tokens_per_second": 846373.2575825447, "lr": 0.0015, "gradient_norm": 0.10500110685825348, "benchmark": false}
+{"step": 12540, "tokens": 6574571520, "loss": 2.793666648864746, "tokens_per_second": 981321.1512611969, "lr": 0.0015, "gradient_norm": 0.0974324643611908, "benchmark": false}
+{"step": 12560, "tokens": 6585057280, "loss": 2.801418495178223, "tokens_per_second": 979704.3133449997, "lr": 0.0015, "gradient_norm": 0.1017347201704979, "benchmark": false}
+{"step": 12580, "tokens": 6595543040, "loss": 2.8266735076904297, "tokens_per_second": 988133.5318305237, "lr": 0.0015, "gradient_norm": 0.11285485327243805, "benchmark": false}
+{"step": 12600, "tokens": 6606028800, "loss": 2.798064041137695, "tokens_per_second": 968544.6581334989, "lr": 0.0015, "gradient_norm": 0.12008693069219589, "benchmark": false}
+{"step": 12620, "tokens": 6616514560, "loss": 2.779056167602539, "tokens_per_second": 974246.924298135, "lr": 0.0015, "gradient_norm": 0.10292800515890121, "benchmark": false}