Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -424,3 +424,7 @@
 {"step": 8480, "tokens": 4445962240, "loss": 2.845271682739258, "tokens_per_second": 996521.2859972045, "lr": 0.0015, "gradient_norm": 0.0774340033531189, "benchmark": false}
 {"step": 8500, "tokens": 4456448000, "loss": 2.8363792419433596, "tokens_per_second": 987704.7458042043, "lr": 0.0015, "gradient_norm": 0.07551994174718857, "benchmark": false}
 {"step": 8520, "tokens": 4466933760, "loss": 2.87451286315918, "tokens_per_second": 831144.4913827988, "lr": 0.0015, "gradient_norm": 0.07500536739826202, "benchmark": false}
+{"step": 8540, "tokens": 4477419520, "loss": 2.8735151290893555, "tokens_per_second": 980860.3832116079, "lr": 0.0015, "gradient_norm": 0.07712480425834656, "benchmark": false}
+{"step": 8560, "tokens": 4487905280, "loss": 2.8743804931640624, "tokens_per_second": 989434.2025171588, "lr": 0.0015, "gradient_norm": 0.07578735798597336, "benchmark": false}
+{"step": 8580, "tokens": 4498391040, "loss": 2.8719303131103517, "tokens_per_second": 996385.3052255794, "lr": 0.0015, "gradient_norm": 0.07415588945150375, "benchmark": false}
+{"step": 8600, "tokens": 4508876800, "loss": 2.8513208389282227, "tokens_per_second": 817336.110214322, "lr": 0.0015, "gradient_norm": 0.07281629741191864, "benchmark": false}