Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -1374,3 +1374,8 @@
 {"step": 27480, "tokens": 14407434240, "loss": 2.723668670654297, "tokens_per_second": 980042.4707011795, "lr": 0.0015, "gradient_norm": 0.17610988020896912, "benchmark": false}
 {"step": 27500, "tokens": 14417920000, "loss": 2.697346496582031, "tokens_per_second": 974398.2166323537, "lr": 0.0015, "gradient_norm": 0.17268800735473633, "benchmark": false}
 {"step": 27520, "tokens": 14428405760, "loss": 2.736563873291016, "tokens_per_second": 848288.5264664899, "lr": 0.0015, "gradient_norm": 0.17833539843559265, "benchmark": false}
+{"step": 27540, "tokens": 14438891520, "loss": 2.693071174621582, "tokens_per_second": 970698.4969644077, "lr": 0.0015, "gradient_norm": 0.20651768147945404, "benchmark": false}
+{"step": 27560, "tokens": 14449377280, "loss": 2.7164464950561524, "tokens_per_second": 979064.8373483644, "lr": 0.0015, "gradient_norm": 0.18882517516613007, "benchmark": false}
+{"step": 27580, "tokens": 14459863040, "loss": 2.731233596801758, "tokens_per_second": 980660.6930201402, "lr": 0.0015, "gradient_norm": 0.18132050335407257, "benchmark": false}
+{"step": 27600, "tokens": 14470348800, "loss": 2.7476808547973635, "tokens_per_second": 979811.8536908344, "lr": 0.0015, "gradient_norm": 0.17652148008346558, "benchmark": false}
+{"step": 27620, "tokens": 14480834560, "loss": 2.742958641052246, "tokens_per_second": 968153.1947297973, "lr": 0.0015, "gradient_norm": 0.17978568375110626, "benchmark": false}