Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -796,3 +796,5 @@
 {"step": 15920, "tokens": 8346664960, "loss": 2.741182327270508, "tokens_per_second": 990415.794020369, "lr": 0.0015, "gradient_norm": 0.13643869757652283, "benchmark": false}
 {"step": 15940, "tokens": 8357150720, "loss": 2.7580354690551756, "tokens_per_second": 987489.3635660079, "lr": 0.0015, "gradient_norm": 0.12679877877235413, "benchmark": false}
 {"step": 15960, "tokens": 8367636480, "loss": 2.7853540420532226, "tokens_per_second": 988360.838972838, "lr": 0.0015, "gradient_norm": 0.12756744027137756, "benchmark": false}
+{"step": 15980, "tokens": 8378122240, "loss": 2.7640474319458006, "tokens_per_second": 981558.347332284, "lr": 0.0015, "gradient_norm": 0.12450078874826431, "benchmark": false}
+{"step": 16000, "tokens": 8388608000, "loss": 2.765132522583008, "tokens_per_second": 986364.2098637007, "lr": 0.0015, "gradient_norm": 0.11989699304103851, "benchmark": false}

reports/run/validation.jsonl

@@ -13,3 +13,4 @@
 {"step": 13000, "source_validation_loss": {"fineweb_edu": 2.9597086906433105, "dclm_edu": 2.994127035140991, "cosmopedia": 2.02358341217041, "finemath": 2.192626714706421}}
 {"step": 14000, "source_validation_loss": {"fineweb_edu": 2.948702096939087, "dclm_edu": 2.9865922927856445, "cosmopedia": 2.0160582065582275, "finemath": 2.18440580368042}}
 {"step": 15000, "source_validation_loss": {"fineweb_edu": 2.9435317516326904, "dclm_edu": 2.9802112579345703, "cosmopedia": 2.0042612552642822, "finemath": 2.18125581741333}}
+{"step": 16000, "source_validation_loss": {"fineweb_edu": 2.9361798763275146, "dclm_edu": 2.9739556312561035, "cosmopedia": 1.9942559003829956, "finemath": 2.171142578125}}