Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -324,3 +324,7 @@
 {"step": 6480, "tokens": 3397386240, "loss": 2.918135643005371, "tokens_per_second": 981682.6743647555, "lr": 0.0015, "gradient_norm": 0.06573039293289185, "benchmark": false}
 {"step": 6500, "tokens": 3407872000, "loss": 2.922238349914551, "tokens_per_second": 967116.1633914592, "lr": 0.0015, "gradient_norm": 0.0701269879937172, "benchmark": false}
 {"step": 6520, "tokens": 3418357760, "loss": 2.8830898284912108, "tokens_per_second": 851558.9785967895, "lr": 0.0015, "gradient_norm": 0.07001926749944687, "benchmark": false}
+{"step": 6540, "tokens": 3428843520, "loss": 2.8867376327514647, "tokens_per_second": 980962.7329971936, "lr": 0.0015, "gradient_norm": 0.06727704405784607, "benchmark": false}
+{"step": 6560, "tokens": 3439329280, "loss": 2.8744361877441404, "tokens_per_second": 811605.0280486693, "lr": 0.0015, "gradient_norm": 0.0776929259300232, "benchmark": false}
+{"step": 6580, "tokens": 3449815040, "loss": 2.8729530334472657, "tokens_per_second": 988722.7496293134, "lr": 0.0015, "gradient_norm": 0.07295847684144974, "benchmark": false}
+{"step": 6600, "tokens": 3460300800, "loss": 2.87994327545166, "tokens_per_second": 985837.3399777771, "lr": 0.0015, "gradient_norm": 0.08181548863649368, "benchmark": false}