Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -223,3 +223,8 @@
 {"step": 4460, "tokens": 2338324480, "loss": 2.952668571472168, "tokens_per_second": 977774.70016757, "lr": 0.0015, "gradient_norm": 0.10796493291854858, "benchmark": false}
 {"step": 4480, "tokens": 2348810240, "loss": 2.937723922729492, "tokens_per_second": 975644.4070345019, "lr": 0.0015, "gradient_norm": 0.10837341099977493, "benchmark": false}
 {"step": 4500, "tokens": 2359296000, "loss": 2.9822259902954102, "tokens_per_second": 972916.4437882437, "lr": 0.0015, "gradient_norm": 0.10424689948558807, "benchmark": false}
+{"step": 4520, "tokens": 2369781760, "loss": 2.9843788146972656, "tokens_per_second": 833079.9661412367, "lr": 0.0015, "gradient_norm": 0.10634101182222366, "benchmark": false}
+{"step": 4540, "tokens": 2380267520, "loss": 2.9769990921020506, "tokens_per_second": 985558.8131431809, "lr": 0.0015, "gradient_norm": 0.1044483482837677, "benchmark": false}
+{"step": 4560, "tokens": 2390753280, "loss": 2.969675827026367, "tokens_per_second": 983421.0391750208, "lr": 0.0015, "gradient_norm": 0.08892372250556946, "benchmark": false}
+{"step": 4580, "tokens": 2401239040, "loss": 2.961945915222168, "tokens_per_second": 969252.2172794534, "lr": 0.0015, "gradient_norm": 0.10147012025117874, "benchmark": false}
+{"step": 4600, "tokens": 2411724800, "loss": 2.9258108139038086, "tokens_per_second": 982775.4198475868, "lr": 0.0015, "gradient_norm": 0.09514818340539932, "benchmark": false}