Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -998,3 +998,8 @@
 {"step": 19960, "tokens": 10464788480, "loss": 2.759548377990723, "tokens_per_second": 985718.732058168, "lr": 0.0015, "gradient_norm": 0.17241807281970978, "benchmark": false}
 {"step": 19980, "tokens": 10475274240, "loss": 2.765959548950195, "tokens_per_second": 984693.2569846577, "lr": 0.0015, "gradient_norm": 0.15740971267223358, "benchmark": false}
 {"step": 20000, "tokens": 10485760000, "loss": 2.7601818084716796, "tokens_per_second": 837521.6981495382, "lr": 0.0015, "gradient_norm": 0.140048548579216, "benchmark": false}
+{"step": 20020, "tokens": 10496245760, "loss": 2.7482187271118166, "tokens_per_second": 815863.3869617886, "lr": 0.0015, "gradient_norm": 0.14850586652755737, "benchmark": false}
+{"step": 20040, "tokens": 10506731520, "loss": 2.7507583618164064, "tokens_per_second": 982837.1355751838, "lr": 0.0015, "gradient_norm": 0.15282844007015228, "benchmark": false}
+{"step": 20060, "tokens": 10517217280, "loss": 2.7343465805053713, "tokens_per_second": 996236.5602341059, "lr": 0.0015, "gradient_norm": 0.1375744342803955, "benchmark": false}
+{"step": 20080, "tokens": 10527703040, "loss": 2.760054588317871, "tokens_per_second": 844460.4750382531, "lr": 0.0015, "gradient_norm": 0.16157233715057373, "benchmark": false}
+{"step": 20100, "tokens": 10538188800, "loss": 2.7180532455444335, "tokens_per_second": 982018.1308888848, "lr": 0.0015, "gradient_norm": 0.14448480308055878, "benchmark": false}