Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -923,3 +923,8 @@
 {"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}
 {"step": 18500, "tokens": 9699328000, "loss": 2.7920162200927736, "tokens_per_second": 981286.7393001046, "lr": 0.0015, "gradient_norm": 0.135093092918396, "benchmark": false}
+{"step": 18520, "tokens": 9709813760, "loss": 2.7670785903930666, "tokens_per_second": 856762.899164113, "lr": 0.0015, "gradient_norm": 0.1572120636701584, "benchmark": false}
+{"step": 18540, "tokens": 9720299520, "loss": 2.7770042419433594, "tokens_per_second": 984636.7704320224, "lr": 0.0015, "gradient_norm": 0.13100825250148773, "benchmark": false}
+{"step": 18560, "tokens": 9730785280, "loss": 2.7588457107543944, "tokens_per_second": 989975.3800258805, "lr": 0.0015, "gradient_norm": 0.1295267790555954, "benchmark": false}
+{"step": 18580, "tokens": 9741271040, "loss": 2.7459754943847656, "tokens_per_second": 982047.222593462, "lr": 0.0015, "gradient_norm": 0.13318170607089996, "benchmark": false}
+{"step": 18600, "tokens": 9751756800, "loss": 2.7676055908203123, "tokens_per_second": 988762.5051151864, "lr": 0.0015, "gradient_norm": 0.1925567090511322, "benchmark": false}