Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -96,3 +96,5 @@
 {"step": 1920, "tokens": 1006632960, "loss": 3.2531501770019533, "tokens_per_second": 976722.1636101804, "lr": 0.0014399999999999999, "gradient_norm": 0.2506352961063385, "benchmark": false}
 {"step": 1940, "tokens": 1017118720, "loss": 3.2249679565429688, "tokens_per_second": 988657.7397372127, "lr": 0.001455, "gradient_norm": 0.2712062895298004, "benchmark": false}
 {"step": 1960, "tokens": 1027604480, "loss": 3.238423156738281, "tokens_per_second": 983368.2203098513, "lr": 0.00147, "gradient_norm": 0.23316670954227448, "benchmark": false}
+{"step": 1980, "tokens": 1038090240, "loss": 3.1997447967529298, "tokens_per_second": 973767.7154982816, "lr": 0.001485, "gradient_norm": 0.28141582012176514, "benchmark": false}
+{"step": 2000, "tokens": 1048576000, "loss": 3.2284637451171876, "tokens_per_second": 976823.1508784221, "lr": 0.0015, "gradient_norm": 0.2324972003698349, "benchmark": false}

reports/run/validation.jsonl

@@ -1 +1,2 @@
 {"step": 1000, "source_validation_loss": {"fineweb_edu": 3.6357946395874023, "dclm_edu": 3.6318132877349854, "cosmopedia": 2.9174532890319824, "finemath": 2.940213680267334}}
+{"step": 2000, "source_validation_loss": {"fineweb_edu": 3.3628950119018555, "dclm_edu": 3.3812646865844727, "cosmopedia": 2.551811456680298, "finemath": 2.6517066955566406}}