Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -824,3 +824,8 @@
 {"step": 16480, "tokens": 8640266240, "loss": 2.756893539428711, "tokens_per_second": 994236.4498590342, "lr": 0.0015, "gradient_norm": 0.13825723528862, "benchmark": false}
 {"step": 16500, "tokens": 8650752000, "loss": 2.751681327819824, "tokens_per_second": 819637.2886612634, "lr": 0.0015, "gradient_norm": 0.12679840624332428, "benchmark": false}
 {"step": 16520, "tokens": 8661237760, "loss": 2.7873161315917967, "tokens_per_second": 860862.4364112108, "lr": 0.0015, "gradient_norm": 0.1185576468706131, "benchmark": false}
+{"step": 16540, "tokens": 8671723520, "loss": 2.7667850494384765, "tokens_per_second": 998642.0541059768, "lr": 0.0015, "gradient_norm": 0.1283944547176361, "benchmark": false}
+{"step": 16560, "tokens": 8682209280, "loss": 2.7945037841796876, "tokens_per_second": 978602.128790704, "lr": 0.0015, "gradient_norm": 0.12816500663757324, "benchmark": false}
+{"step": 16580, "tokens": 8692695040, "loss": 2.767159652709961, "tokens_per_second": 991293.4213889013, "lr": 0.0015, "gradient_norm": 0.12242449820041656, "benchmark": false}
+{"step": 16600, "tokens": 8703180800, "loss": 2.797515869140625, "tokens_per_second": 993813.742701105, "lr": 0.0015, "gradient_norm": 0.12738817930221558, "benchmark": false}
+{"step": 16620, "tokens": 8713666560, "loss": 2.738190269470215, "tokens_per_second": 990569.5727881185, "lr": 0.0015, "gradient_norm": 0.14329010248184204, "benchmark": false}