Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -1396,3 +1396,5 @@
 {"step": 27920, "tokens": 14638120960, "loss": 2.7415027618408203, "tokens_per_second": 973208.4045714834, "lr": 0.0015, "gradient_norm": 0.18788912892341614, "benchmark": false}
 {"step": 27940, "tokens": 14648606720, "loss": 2.699085998535156, "tokens_per_second": 967242.1150041102, "lr": 0.0015, "gradient_norm": 0.1698664277791977, "benchmark": false}
 {"step": 27960, "tokens": 14659092480, "loss": 2.710794448852539, "tokens_per_second": 971998.9333785562, "lr": 0.0015, "gradient_norm": 0.15751716494560242, "benchmark": false}
+{"step": 27980, "tokens": 14669578240, "loss": 2.707795333862305, "tokens_per_second": 966638.7383064849, "lr": 0.0015, "gradient_norm": 0.1895906627178192, "benchmark": false}
+{"step": 28000, "tokens": 14680064000, "loss": 2.7293113708496093, "tokens_per_second": 983400.759093925, "lr": 0.0015, "gradient_norm": 0.20459537208080292, "benchmark": false}

reports/run/validation.jsonl

@@ -25,3 +25,4 @@
 {"step": 25000, "source_validation_loss": {"fineweb_edu": 2.8956382274627686, "dclm_edu": 2.936393976211548, "cosmopedia": 1.9424511194229126, "finemath": 2.1288459300994873}}
 {"step": 26000, "source_validation_loss": {"fineweb_edu": 2.893357038497925, "dclm_edu": 2.934877634048462, "cosmopedia": 1.9365743398666382, "finemath": 2.1286120414733887}}
 {"step": 27000, "source_validation_loss": {"fineweb_edu": 2.890822649002075, "dclm_edu": 2.9303057193756104, "cosmopedia": 1.936927318572998, "finemath": 2.1255106925964355}}
+{"step": 28000, "source_validation_loss": {"fineweb_edu": 2.887849807739258, "dclm_edu": 2.930004119873047, "cosmopedia": 1.933445692062378, "finemath": 2.1189560890197754}}