Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -446,3 +446,5 @@
 {"step": 8920, "tokens": 4676648960, "loss": 2.860857391357422, "tokens_per_second": 977166.5448684128, "lr": 0.0015, "gradient_norm": 0.08019018173217773, "benchmark": false}
 {"step": 8940, "tokens": 4687134720, "loss": 2.8259984970092775, "tokens_per_second": 973040.8571244124, "lr": 0.0015, "gradient_norm": 0.0780382826924324, "benchmark": false}
 {"step": 8960, "tokens": 4697620480, "loss": 2.8529546737670897, "tokens_per_second": 969905.4493514767, "lr": 0.0015, "gradient_norm": 0.08122095465660095, "benchmark": false}
+{"step": 8980, "tokens": 4708106240, "loss": 2.840422821044922, "tokens_per_second": 972420.6085403181, "lr": 0.0015, "gradient_norm": 0.0813227966427803, "benchmark": false}
+{"step": 9000, "tokens": 4718592000, "loss": 2.819308853149414, "tokens_per_second": 984628.671317153, "lr": 0.0015, "gradient_norm": 0.07416989654302597, "benchmark": false}

reports/run/validation.jsonl

@@ -6,3 +6,4 @@
 {"step": 6000, "source_validation_loss": {"fineweb_edu": 3.0642285346984863, "dclm_edu": 3.0997276306152344, "cosmopedia": 2.1647276878356934, "finemath": 2.31894588470459}}
 {"step": 7000, "source_validation_loss": {"fineweb_edu": 3.0427091121673584, "dclm_edu": 3.072791576385498, "cosmopedia": 2.1341612339019775, "finemath": 2.284400701522827}}
 {"step": 8000, "source_validation_loss": {"fineweb_edu": 3.022184133529663, "dclm_edu": 3.0521011352539062, "cosmopedia": 2.1067497730255127, "finemath": 2.2722864151000977}}
+{"step": 9000, "source_validation_loss": {"fineweb_edu": 3.0031962394714355, "dclm_edu": 3.0357918739318848, "cosmopedia": 2.0862884521484375, "finemath": 2.2452752590179443}}