Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -873,3 +873,7 @@
 {"step": 17460, "tokens": 9154068480, "loss": 2.756350517272949, "tokens_per_second": 963971.2649856538, "lr": 0.0015, "gradient_norm": 0.13418878614902496, "benchmark": false}
 {"step": 17480, "tokens": 9164554240, "loss": 2.7274747848510743, "tokens_per_second": 962931.4099102088, "lr": 0.0015, "gradient_norm": 0.1287911832332611, "benchmark": false}
 {"step": 17500, "tokens": 9175040000, "loss": 2.775168228149414, "tokens_per_second": 968762.2304840548, "lr": 0.0015, "gradient_norm": 0.13094007968902588, "benchmark": false}
+{"step": 17520, "tokens": 9185525760, "loss": 2.7674982070922853, "tokens_per_second": 851806.8391884612, "lr": 0.0015, "gradient_norm": 0.15391556918621063, "benchmark": false}
+{"step": 17540, "tokens": 9196011520, "loss": 2.7649585723876955, "tokens_per_second": 978741.8174425006, "lr": 0.0015, "gradient_norm": 0.1797247678041458, "benchmark": false}
+{"step": 17560, "tokens": 9206497280, "loss": 2.772072601318359, "tokens_per_second": 985750.2007544704, "lr": 0.0015, "gradient_norm": 0.13320325314998627, "benchmark": false}
+{"step": 17580, "tokens": 9216983040, "loss": 2.7577545166015627, "tokens_per_second": 953086.8095424904, "lr": 0.0015, "gradient_norm": 0.1323719322681427, "benchmark": false}