Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -1398,3 +1398,8 @@
 {"step": 27960, "tokens": 14659092480, "loss": 2.710794448852539, "tokens_per_second": 971998.9333785562, "lr": 0.0015, "gradient_norm": 0.15751716494560242, "benchmark": false}
 {"step": 27980, "tokens": 14669578240, "loss": 2.707795333862305, "tokens_per_second": 966638.7383064849, "lr": 0.0015, "gradient_norm": 0.1895906627178192, "benchmark": false}
 {"step": 28000, "tokens": 14680064000, "loss": 2.7293113708496093, "tokens_per_second": 983400.759093925, "lr": 0.0015, "gradient_norm": 0.20459537208080292, "benchmark": false}
+{"step": 28020, "tokens": 14690549760, "loss": 2.7033313751220702, "tokens_per_second": 733110.2691602296, "lr": 0.0015, "gradient_norm": 0.16380923986434937, "benchmark": false}
+{"step": 28040, "tokens": 14701035520, "loss": 2.7154237747192385, "tokens_per_second": 982207.2376210534, "lr": 0.0015, "gradient_norm": 0.19475558400154114, "benchmark": false}
+{"step": 28060, "tokens": 14711521280, "loss": 2.681853485107422, "tokens_per_second": 945412.9039389335, "lr": 0.0015, "gradient_norm": 0.2263222336769104, "benchmark": false}
+{"step": 28080, "tokens": 14722007040, "loss": 2.7147281646728514, "tokens_per_second": 958340.254903997, "lr": 0.0015, "gradient_norm": 0.17466644942760468, "benchmark": false}
+{"step": 28100, "tokens": 14732492800, "loss": 2.6936927795410157, "tokens_per_second": 929234.3664731835, "lr": 0.0015, "gradient_norm": 0.19458916783332825, "benchmark": false}