Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -1298,3 +1298,8 @@
 {"step": 25960, "tokens": 13610516480, "loss": 2.708211135864258, "tokens_per_second": 854841.7415926263, "lr": 0.0015, "gradient_norm": 0.17577874660491943, "benchmark": false}
 {"step": 25980, "tokens": 13621002240, "loss": 2.712994384765625, "tokens_per_second": 973681.908757307, "lr": 0.0015, "gradient_norm": 0.1609300822019577, "benchmark": false}
 {"step": 26000, "tokens": 13631488000, "loss": 2.724579620361328, "tokens_per_second": 967542.1501508587, "lr": 0.0015, "gradient_norm": 0.1633884310722351, "benchmark": false}
+{"step": 26020, "tokens": 13641973760, "loss": 2.7292980194091796, "tokens_per_second": 818497.4700031824, "lr": 0.0015, "gradient_norm": 0.16491888463497162, "benchmark": false}
+{"step": 26040, "tokens": 13652459520, "loss": 2.7431610107421873, "tokens_per_second": 986377.2939282341, "lr": 0.0015, "gradient_norm": 0.16621766984462738, "benchmark": false}
+{"step": 26060, "tokens": 13662945280, "loss": 2.7311733245849608, "tokens_per_second": 936785.07679417, "lr": 0.0015, "gradient_norm": 0.1704351156949997, "benchmark": false}
+{"step": 26080, "tokens": 13673431040, "loss": 2.7357757568359373, "tokens_per_second": 973855.9108424016, "lr": 0.0015, "gradient_norm": 0.18583223223686218, "benchmark": false}
+{"step": 26100, "tokens": 13683916800, "loss": 2.7040576934814453, "tokens_per_second": 969912.2554433204, "lr": 0.0015, "gradient_norm": 0.1684197634458542, "benchmark": false}