Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -173,3 +173,8 @@
 {"step": 3460, "tokens": 1814036480, "loss": 3.0570775985717775, "tokens_per_second": 969805.7895285914, "lr": 0.0015, "gradient_norm": 0.15748904645442963, "benchmark": false}
 {"step": 3480, "tokens": 1824522240, "loss": 3.0283912658691405, "tokens_per_second": 969750.7373291658, "lr": 0.0015, "gradient_norm": 0.17056794464588165, "benchmark": false}
 {"step": 3500, "tokens": 1835008000, "loss": 3.034389114379883, "tokens_per_second": 971371.564367871, "lr": 0.0015, "gradient_norm": 0.13611756265163422, "benchmark": false}
+{"step": 3520, "tokens": 1845493760, "loss": 3.0094444274902346, "tokens_per_second": 845365.5708885989, "lr": 0.0015, "gradient_norm": 0.12842446565628052, "benchmark": false}
+{"step": 3540, "tokens": 1855979520, "loss": 3.02652587890625, "tokens_per_second": 993343.9287551965, "lr": 0.0015, "gradient_norm": 0.12900671362876892, "benchmark": false}
+{"step": 3560, "tokens": 1866465280, "loss": 3.005375862121582, "tokens_per_second": 914666.6063207821, "lr": 0.0015, "gradient_norm": 0.1524912416934967, "benchmark": false}
+{"step": 3580, "tokens": 1876951040, "loss": 3.01119441986084, "tokens_per_second": 979113.9151245613, "lr": 0.0015, "gradient_norm": 0.1300109326839447, "benchmark": false}
+{"step": 3600, "tokens": 1887436800, "loss": 2.971750831604004, "tokens_per_second": 974042.0355774015, "lr": 0.0015, "gradient_norm": 0.15577509999275208, "benchmark": false}