Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -1323,3 +1323,7 @@
 {"step": 26460, "tokens": 13872660480, "loss": 2.729097366333008, "tokens_per_second": 992203.155856663, "lr": 0.0015, "gradient_norm": 0.16647778451442719, "benchmark": false}
 {"step": 26480, "tokens": 13883146240, "loss": 2.7267486572265627, "tokens_per_second": 988070.7718164688, "lr": 0.0015, "gradient_norm": 0.1715329885482788, "benchmark": false}
 {"step": 26500, "tokens": 13893632000, "loss": 2.7060214996337892, "tokens_per_second": 991145.660072251, "lr": 0.0015, "gradient_norm": 0.1872853934764862, "benchmark": false}
+{"step": 26520, "tokens": 13904117760, "loss": 2.736189842224121, "tokens_per_second": 856868.2480380794, "lr": 0.0015, "gradient_norm": 0.15069472789764404, "benchmark": false}
+{"step": 26540, "tokens": 13914603520, "loss": 2.682301902770996, "tokens_per_second": 978042.5496547779, "lr": 0.0015, "gradient_norm": 0.17297181487083435, "benchmark": false}
+{"step": 26560, "tokens": 13925089280, "loss": 2.7173273086547853, "tokens_per_second": 982403.8772163352, "lr": 0.0015, "gradient_norm": 0.18856073915958405, "benchmark": false}
+{"step": 26580, "tokens": 13935575040, "loss": 2.726406478881836, "tokens_per_second": 992744.8377322793, "lr": 0.0015, "gradient_norm": 0.2031981199979782, "benchmark": false}