Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -1248,3 +1248,8 @@
 {"step": 24960, "tokens": 13086228480, "loss": 2.6991626739501955, "tokens_per_second": 939572.5153784158, "lr": 0.0015, "gradient_norm": 0.1650470644235611, "benchmark": false}
 {"step": 24980, "tokens": 13096714240, "loss": 2.714500617980957, "tokens_per_second": 984047.7388723607, "lr": 0.0015, "gradient_norm": 0.1731661707162857, "benchmark": false}
 {"step": 25000, "tokens": 13107200000, "loss": 2.7156171798706055, "tokens_per_second": 999762.4758471423, "lr": 0.0015, "gradient_norm": 0.17807134985923767, "benchmark": false}
+{"step": 25020, "tokens": 13117685760, "loss": 2.6884231567382812, "tokens_per_second": 727366.9593482548, "lr": 0.0015, "gradient_norm": 0.16767993569374084, "benchmark": false}
+{"step": 25040, "tokens": 13128171520, "loss": 2.7099437713623047, "tokens_per_second": 994332.6324797974, "lr": 0.0015, "gradient_norm": 0.17758285999298096, "benchmark": false}
+{"step": 25060, "tokens": 13138657280, "loss": 2.700517463684082, "tokens_per_second": 984705.5389486365, "lr": 0.0015, "gradient_norm": 0.23347370326519012, "benchmark": false}
+{"step": 25080, "tokens": 13149143040, "loss": 2.6942159652709963, "tokens_per_second": 985476.2811881976, "lr": 0.0015, "gradient_norm": 0.16518275439739227, "benchmark": false}
+{"step": 25100, "tokens": 13159628800, "loss": 2.7206411361694336, "tokens_per_second": 976840.2746375389, "lr": 0.0015, "gradient_norm": 0.167807474732399, "benchmark": false}