Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -599,3 +599,7 @@
 {"step": 11980, "tokens": 6280970240, "loss": 2.8234445571899416, "tokens_per_second": 847536.274677034, "lr": 0.0015, "gradient_norm": 0.10739849507808685, "benchmark": false}
 {"step": 12000, "tokens": 6291456000, "loss": 2.754237174987793, "tokens_per_second": 987605.9263153236, "lr": 0.0015, "gradient_norm": 0.10888316482305527, "benchmark": false}
 {"step": 12020, "tokens": 6301941760, "loss": 2.8089101791381834, "tokens_per_second": 822342.5921110236, "lr": 0.0015, "gradient_norm": 0.10958659648895264, "benchmark": false}
+{"step": 12040, "tokens": 6312427520, "loss": 2.8103054046630858, "tokens_per_second": 979209.831793374, "lr": 0.0015, "gradient_norm": 0.08929789811372757, "benchmark": false}
+{"step": 12060, "tokens": 6322913280, "loss": 2.8015066146850587, "tokens_per_second": 983909.7219845104, "lr": 0.0015, "gradient_norm": 0.0960715189576149, "benchmark": false}
+{"step": 12080, "tokens": 6333399040, "loss": 2.789211654663086, "tokens_per_second": 986576.222179265, "lr": 0.0015, "gradient_norm": 0.09555419534444809, "benchmark": false}
+{"step": 12100, "tokens": 6343884800, "loss": 2.779342842102051, "tokens_per_second": 839032.7184058921, "lr": 0.0015, "gradient_norm": 0.11269618570804596, "benchmark": false}