Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -473,3 +473,8 @@
 {"step": 9460, "tokens": 4959764480, "loss": 2.836428642272949, "tokens_per_second": 984975.3316094908, "lr": 0.0015, "gradient_norm": 0.08083634823560715, "benchmark": false}
 {"step": 9480, "tokens": 4970250240, "loss": 2.838408279418945, "tokens_per_second": 993077.121682915, "lr": 0.0015, "gradient_norm": 0.08504550904035568, "benchmark": false}
 {"step": 9500, "tokens": 4980736000, "loss": 2.847154998779297, "tokens_per_second": 954012.1774107493, "lr": 0.0015, "gradient_norm": 0.08038152009248734, "benchmark": false}
+{"step": 9520, "tokens": 4991221760, "loss": 2.8305084228515627, "tokens_per_second": 840798.1068664312, "lr": 0.0015, "gradient_norm": 0.085055410861969, "benchmark": false}
+{"step": 9540, "tokens": 5001707520, "loss": 2.833781051635742, "tokens_per_second": 982512.4793784562, "lr": 0.0015, "gradient_norm": 0.08446768671274185, "benchmark": false}
+{"step": 9560, "tokens": 5012193280, "loss": 2.8237529754638673, "tokens_per_second": 989944.2442785497, "lr": 0.0015, "gradient_norm": 0.0901399552822113, "benchmark": false}
+{"step": 9580, "tokens": 5022679040, "loss": 2.8359748840332033, "tokens_per_second": 984694.8421158455, "lr": 0.0015, "gradient_norm": 0.08997666090726852, "benchmark": false}
+{"step": 9600, "tokens": 5033164800, "loss": 2.862395668029785, "tokens_per_second": 989213.3500881932, "lr": 0.0015, "gradient_norm": 0.07941855490207672, "benchmark": false}