Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -849,3 +849,7 @@
 {"step": 16980, "tokens": 8902410240, "loss": 2.764535331726074, "tokens_per_second": 975047.8654111625, "lr": 0.0015, "gradient_norm": 0.1423749327659607, "benchmark": false}
 {"step": 17000, "tokens": 8912896000, "loss": 2.7730123519897463, "tokens_per_second": 981977.874173138, "lr": 0.0015, "gradient_norm": 0.13165350258350372, "benchmark": false}
 {"step": 17020, "tokens": 8923381760, "loss": 2.743798828125, "tokens_per_second": 818942.7578559162, "lr": 0.0015, "gradient_norm": 0.13705229759216309, "benchmark": false}
+{"step": 17040, "tokens": 8933867520, "loss": 2.7651670455932615, "tokens_per_second": 978044.6029568557, "lr": 0.0015, "gradient_norm": 0.14534282684326172, "benchmark": false}
+{"step": 17060, "tokens": 8944353280, "loss": 2.7328630447387696, "tokens_per_second": 987639.7265434645, "lr": 0.0015, "gradient_norm": 0.12985369563102722, "benchmark": false}
+{"step": 17080, "tokens": 8954839040, "loss": 2.7397464752197265, "tokens_per_second": 982369.9370642095, "lr": 0.0015, "gradient_norm": 0.13668425381183624, "benchmark": false}
+{"step": 17100, "tokens": 8965324800, "loss": 2.7487125396728516, "tokens_per_second": 981721.9066029039, "lr": 0.0015, "gradient_norm": 0.12753145396709442, "benchmark": false}