Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -124,3 +124,8 @@
 {"step": 2480, "tokens": 1300234240, "loss": 3.116751861572266, "tokens_per_second": 983559.1611577123, "lr": 0.0015, "gradient_norm": 0.22678160667419434, "benchmark": false}
 {"step": 2500, "tokens": 1310720000, "loss": 3.1567005157470702, "tokens_per_second": 820689.2260704602, "lr": 0.0015, "gradient_norm": 0.2704465389251709, "benchmark": false}
 {"step": 2520, "tokens": 1321205760, "loss": 3.128544235229492, "tokens_per_second": 848509.4357302934, "lr": 0.0015, "gradient_norm": 0.23646458983421326, "benchmark": false}
+{"step": 2540, "tokens": 1331691520, "loss": 3.126090431213379, "tokens_per_second": 988628.4239650108, "lr": 0.0015, "gradient_norm": 0.22039943933486938, "benchmark": false}
+{"step": 2560, "tokens": 1342177280, "loss": 3.120811653137207, "tokens_per_second": 989099.8075437183, "lr": 0.0015, "gradient_norm": 0.26951202750205994, "benchmark": false}
+{"step": 2580, "tokens": 1352663040, "loss": 3.124241065979004, "tokens_per_second": 930455.2500459759, "lr": 0.0015, "gradient_norm": 0.24497447907924652, "benchmark": false}
+{"step": 2600, "tokens": 1363148800, "loss": 3.1559934616088867, "tokens_per_second": 989359.9285415264, "lr": 0.0015, "gradient_norm": 0.2681301236152649, "benchmark": false}
+{"step": 2620, "tokens": 1373634560, "loss": 3.120655632019043, "tokens_per_second": 988014.3404262743, "lr": 0.0015, "gradient_norm": 0.22039465606212616, "benchmark": false}