Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -1098,3 +1098,8 @@
 {"step": 21960, "tokens": 11513364480, "loss": 2.7490068435668946, "tokens_per_second": 938317.9829802986, "lr": 0.0015, "gradient_norm": 0.1540951132774353, "benchmark": false}
 {"step": 21980, "tokens": 11523850240, "loss": 2.7099571228027344, "tokens_per_second": 972480.1144068335, "lr": 0.0015, "gradient_norm": 0.14107529819011688, "benchmark": false}
 {"step": 22000, "tokens": 11534336000, "loss": 2.7759721755981444, "tokens_per_second": 971667.9170136879, "lr": 0.0015, "gradient_norm": 0.17356452345848083, "benchmark": false}
+{"step": 22020, "tokens": 11544821760, "loss": 2.7497358322143555, "tokens_per_second": 724434.4823868582, "lr": 0.0015, "gradient_norm": 0.18714021146297455, "benchmark": false}
+{"step": 22040, "tokens": 11555307520, "loss": 2.7553672790527344, "tokens_per_second": 998087.4912387363, "lr": 0.0015, "gradient_norm": 0.17349259555339813, "benchmark": false}
+{"step": 22060, "tokens": 11565793280, "loss": 2.7409873962402345, "tokens_per_second": 986650.8122801217, "lr": 0.0015, "gradient_norm": 0.17116031050682068, "benchmark": false}
+{"step": 22080, "tokens": 11576279040, "loss": 2.7698875427246095, "tokens_per_second": 989118.4513079759, "lr": 0.0015, "gradient_norm": 0.14208100736141205, "benchmark": false}
+{"step": 22100, "tokens": 11586764800, "loss": 2.7190452575683595, "tokens_per_second": 977903.7618479038, "lr": 0.0015, "gradient_norm": 0.16330455243587494, "benchmark": false}