Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -647,3 +647,5 @@
 {"step": 12940, "tokens": 6784286720, "loss": 2.809139633178711, "tokens_per_second": 974379.8457599642, "lr": 0.0015, "gradient_norm": 0.11274027824401855, "benchmark": false}
 {"step": 12960, "tokens": 6794772480, "loss": 2.808652877807617, "tokens_per_second": 975053.6049672521, "lr": 0.0015, "gradient_norm": 0.10414215177297592, "benchmark": false}
 {"step": 12980, "tokens": 6805258240, "loss": 2.789379692077637, "tokens_per_second": 975217.6425159012, "lr": 0.0015, "gradient_norm": 0.11947929114103317, "benchmark": false}
+{"step": 13000, "tokens": 6815744000, "loss": 2.766477584838867, "tokens_per_second": 974659.1020516185, "lr": 0.0015, "gradient_norm": 0.10573747754096985, "benchmark": false}
+{"step": 13020, "tokens": 6826229760, "loss": 2.766706657409668, "tokens_per_second": 819334.3486186384, "lr": 0.0015, "gradient_norm": 0.10982190817594528, "benchmark": false}

reports/run/validation.jsonl

@@ -10,3 +10,4 @@
 {"step": 10000, "source_validation_loss": {"fineweb_edu": 2.9892542362213135, "dclm_edu": 3.0245914459228516, "cosmopedia": 2.067286968231201, "finemath": 2.238250970840454}}
 {"step": 11000, "source_validation_loss": {"fineweb_edu": 2.9761719703674316, "dclm_edu": 3.0146288871765137, "cosmopedia": 2.05277943611145, "finemath": 2.2114815711975098}}
 {"step": 12000, "source_validation_loss": {"fineweb_edu": 2.9684195518493652, "dclm_edu": 3.0029680728912354, "cosmopedia": 2.0343093872070312, "finemath": 2.201632261276245}}
+{"step": 13000, "source_validation_loss": {"fineweb_edu": 2.9597086906433105, "dclm_edu": 2.994127035140991, "cosmopedia": 2.02358341217041, "finemath": 2.192626714706421}}