Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -774,3 +774,7 @@
 {"step": 15480, "tokens": 8115978240, "loss": 2.774624824523926, "tokens_per_second": 987296.5932243799, "lr": 0.0015, "gradient_norm": 0.12382978200912476, "benchmark": false}
 {"step": 15500, "tokens": 8126464000, "loss": 2.7601093292236327, "tokens_per_second": 984571.0421259468, "lr": 0.0015, "gradient_norm": 0.1273164451122284, "benchmark": false}
 {"step": 15520, "tokens": 8136949760, "loss": 2.7832685470581056, "tokens_per_second": 861158.9010310016, "lr": 0.0015, "gradient_norm": 0.12241510301828384, "benchmark": false}
+{"step": 15540, "tokens": 8147435520, "loss": 2.786355209350586, "tokens_per_second": 987504.7467624149, "lr": 0.0015, "gradient_norm": 0.13968412578105927, "benchmark": false}
+{"step": 15560, "tokens": 8157921280, "loss": 2.771760177612305, "tokens_per_second": 991712.5755314733, "lr": 0.0015, "gradient_norm": 0.12112805247306824, "benchmark": false}
+{"step": 15580, "tokens": 8168407040, "loss": 2.7709102630615234, "tokens_per_second": 992927.0085300705, "lr": 0.0015, "gradient_norm": 0.12343323230743408, "benchmark": false}
+{"step": 15600, "tokens": 8178892800, "loss": 2.761789321899414, "tokens_per_second": 791803.2874319528, "lr": 0.0015, "gradient_norm": 0.12571533024311066, "benchmark": false}