Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -948,3 +948,7 @@
 {"step": 18960, "tokens": 9940500480, "loss": 2.751300811767578, "tokens_per_second": 977357.1369480103, "lr": 0.0015, "gradient_norm": 0.14042624831199646, "benchmark": false}
 {"step": 18980, "tokens": 9950986240, "loss": 2.731955718994141, "tokens_per_second": 973377.4680011334, "lr": 0.0015, "gradient_norm": 0.13745440542697906, "benchmark": false}
 {"step": 19000, "tokens": 9961472000, "loss": 2.720782661437988, "tokens_per_second": 969674.3917193501, "lr": 0.0015, "gradient_norm": 0.15510135889053345, "benchmark": false}
+{"step": 19020, "tokens": 9971957760, "loss": 2.736471748352051, "tokens_per_second": 827558.4382417401, "lr": 0.0015, "gradient_norm": 0.14572863280773163, "benchmark": false}
+{"step": 19040, "tokens": 9982443520, "loss": 2.753316307067871, "tokens_per_second": 903611.3654674874, "lr": 0.0015, "gradient_norm": 0.12864474952220917, "benchmark": false}
+{"step": 19060, "tokens": 9992929280, "loss": 2.6966789245605467, "tokens_per_second": 983493.5365173507, "lr": 0.0015, "gradient_norm": 0.14442148804664612, "benchmark": false}
+{"step": 19080, "tokens": 10003415040, "loss": 2.747500991821289, "tokens_per_second": 986488.5372464606, "lr": 0.0015, "gradient_norm": 0.15878035128116608, "benchmark": false}