Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -697,3 +697,5 @@
 {"step": 13940, "tokens": 7308574720, "loss": 2.768319320678711, "tokens_per_second": 964042.0161144747, "lr": 0.0015, "gradient_norm": 0.1116306260228157, "benchmark": false}
 {"step": 13960, "tokens": 7319060480, "loss": 2.797551155090332, "tokens_per_second": 975699.4953853759, "lr": 0.0015, "gradient_norm": 0.11029012501239777, "benchmark": false}
 {"step": 13980, "tokens": 7329546240, "loss": 2.7930103302001954, "tokens_per_second": 970508.686897241, "lr": 0.0015, "gradient_norm": 0.12515392899513245, "benchmark": false}
+{"step": 14000, "tokens": 7340032000, "loss": 2.810237503051758, "tokens_per_second": 963901.6917292763, "lr": 0.0015, "gradient_norm": 0.11248497664928436, "benchmark": false}
+{"step": 14020, "tokens": 7350517760, "loss": 2.791335678100586, "tokens_per_second": 813652.9955171734, "lr": 0.0015, "gradient_norm": 0.11959215253591537, "benchmark": false}

reports/run/validation.jsonl

@@ -11,3 +11,4 @@
 {"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}}
+{"step": 14000, "source_validation_loss": {"fineweb_edu": 2.948702096939087, "dclm_edu": 2.9865922927856445, "cosmopedia": 2.0160582065582275, "finemath": 2.18440580368042}}