Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -919,3 +919,6 @@
 {"step": 18380, "tokens": 9636413440, "loss": 2.7541200637817385, "tokens_per_second": 978007.5796573953, "lr": 0.0015, "gradient_norm": 0.13819445669651031, "benchmark": false}
 {"step": 18400, "tokens": 9646899200, "loss": 2.7527652740478517, "tokens_per_second": 979229.8178542908, "lr": 0.0015, "gradient_norm": 0.14924854040145874, "benchmark": false}
 {"step": 18420, "tokens": 9657384960, "loss": 2.7542049407958986, "tokens_per_second": 990100.4783217756, "lr": 0.0015, "gradient_norm": 0.12890124320983887, "benchmark": false}
+{"step": 18440, "tokens": 9667870720, "loss": 2.7615888595581053, "tokens_per_second": 978303.8545806182, "lr": 0.0015, "gradient_norm": 0.12055376172065735, "benchmark": false}
+{"step": 18460, "tokens": 9678356480, "loss": 2.744034767150879, "tokens_per_second": 984021.9975202855, "lr": 0.0015, "gradient_norm": 0.12270098924636841, "benchmark": false}
+{"step": 18480, "tokens": 9688842240, "loss": 2.7206756591796877, "tokens_per_second": 984063.8381700526, "lr": 0.0015, "gradient_norm": 0.14856895804405212, "benchmark": false}