Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -298,3 +298,7 @@
 {"step": 5960, "tokens": 3124756480, "loss": 2.9387432098388673, "tokens_per_second": 978047.3818877105, "lr": 0.0015, "gradient_norm": 0.0761580765247345, "benchmark": false}
 {"step": 5980, "tokens": 3135242240, "loss": 2.8938913345336914, "tokens_per_second": 971113.1983575885, "lr": 0.0015, "gradient_norm": 0.07130406051874161, "benchmark": false}
 {"step": 6000, "tokens": 3145728000, "loss": 2.910635757446289, "tokens_per_second": 968513.8750935235, "lr": 0.0015, "gradient_norm": 0.07274629175662994, "benchmark": false}
+{"step": 6020, "tokens": 3156213760, "loss": 2.9167280197143555, "tokens_per_second": 813333.7822438873, "lr": 0.0015, "gradient_norm": 0.07697819173336029, "benchmark": false}
+{"step": 6040, "tokens": 3166699520, "loss": 2.919988250732422, "tokens_per_second": 988199.3814672184, "lr": 0.0015, "gradient_norm": 0.07092317193746567, "benchmark": false}
+{"step": 6060, "tokens": 3177185280, "loss": 2.8964223861694336, "tokens_per_second": 986418.2932919076, "lr": 0.0015, "gradient_norm": 0.0735328271985054, "benchmark": false}
+{"step": 6080, "tokens": 3187671040, "loss": 2.9418666839599608, "tokens_per_second": 981378.5804716933, "lr": 0.0015, "gradient_norm": 0.09998906403779984, "benchmark": false}