Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -748,3 +748,7 @@
 {"step": 14960, "tokens": 7843348480, "loss": 2.8001480102539062, "tokens_per_second": 989712.792759918, "lr": 0.0015, "gradient_norm": 0.12138649076223373, "benchmark": false}
 {"step": 14980, "tokens": 7853834240, "loss": 2.7922641754150392, "tokens_per_second": 990706.2731206587, "lr": 0.0015, "gradient_norm": 0.11834993213415146, "benchmark": false}
 {"step": 15000, "tokens": 7864320000, "loss": 2.810902976989746, "tokens_per_second": 985215.5444990908, "lr": 0.0015, "gradient_norm": 0.13044802844524384, "benchmark": false}
+{"step": 15020, "tokens": 7874805760, "loss": 2.7654020309448244, "tokens_per_second": 823555.812169737, "lr": 0.0015, "gradient_norm": 0.12740527093410492, "benchmark": false}
+{"step": 15040, "tokens": 7885291520, "loss": 2.800482749938965, "tokens_per_second": 977951.2339340949, "lr": 0.0015, "gradient_norm": 0.11727306246757507, "benchmark": false}
+{"step": 15060, "tokens": 7895777280, "loss": 2.7801792144775392, "tokens_per_second": 984479.5907513736, "lr": 0.0015, "gradient_norm": 0.13086150586605072, "benchmark": false}
+{"step": 15080, "tokens": 7906263040, "loss": 2.7915508270263674, "tokens_per_second": 991798.1438708312, "lr": 0.0015, "gradient_norm": 0.1386176198720932, "benchmark": false}