Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -974,3 +974,8 @@
 {"step": 19480, "tokens": 10213130240, "loss": 2.770153045654297, "tokens_per_second": 985265.7709530981, "lr": 0.0015, "gradient_norm": 0.17231428623199463, "benchmark": false}
 {"step": 19500, "tokens": 10223616000, "loss": 2.7508995056152346, "tokens_per_second": 987447.5489612769, "lr": 0.0015, "gradient_norm": 0.15210653841495514, "benchmark": false}
 {"step": 19520, "tokens": 10234101760, "loss": 2.754379463195801, "tokens_per_second": 868624.3791890546, "lr": 0.0015, "gradient_norm": 0.14836817979812622, "benchmark": false}
+{"step": 19540, "tokens": 10244587520, "loss": 2.7675302505493162, "tokens_per_second": 990896.1461014918, "lr": 0.0015, "gradient_norm": 0.1669021099805832, "benchmark": false}
+{"step": 19560, "tokens": 10255073280, "loss": 2.754825973510742, "tokens_per_second": 942646.142548731, "lr": 0.0015, "gradient_norm": 0.14446082711219788, "benchmark": false}
+{"step": 19580, "tokens": 10265559040, "loss": 2.769974136352539, "tokens_per_second": 987735.7227031054, "lr": 0.0015, "gradient_norm": 0.20880039036273956, "benchmark": false}
+{"step": 19600, "tokens": 10276044800, "loss": 2.716664505004883, "tokens_per_second": 971903.9757540174, "lr": 0.0015, "gradient_norm": 0.1424456536769867, "benchmark": false}
+{"step": 19620, "tokens": 10286530560, "loss": 2.784602737426758, "tokens_per_second": 976148.7554122006, "lr": 0.0015, "gradient_norm": 0.16809336841106415, "benchmark": false}