Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -4,3 +4,6 @@
 {"step": 80, "tokens": 41943040, "loss": 8.688764190673828, "tokens_per_second": 1032255.2836004663, "lr": 6e-05, "gradient_norm": 1.4424148797988892, "benchmark": false}
 {"step": 100, "tokens": 52428800, "loss": 8.105000305175782, "tokens_per_second": 1029399.3815147971, "lr": 7.500000000000001e-05, "gradient_norm": 1.0234349966049194, "benchmark": false}
 {"step": 120, "tokens": 62914560, "loss": 7.616502380371093, "tokens_per_second": 1015709.3869896838, "lr": 8.999999999999999e-05, "gradient_norm": 1.1642669439315796, "benchmark": false}
+{"step": 140, "tokens": 73400320, "loss": 7.291810607910156, "tokens_per_second": 1015522.4437225041, "lr": 0.00010500000000000002, "gradient_norm": 0.8104191422462463, "benchmark": false}
+{"step": 160, "tokens": 83886080, "loss": 7.000241088867187, "tokens_per_second": 1024867.7143696964, "lr": 0.00012, "gradient_norm": 0.8440332412719727, "benchmark": false}
+{"step": 180, "tokens": 94371840, "loss": 6.713601684570312, "tokens_per_second": 1024219.9743449177, "lr": 0.000135, "gradient_norm": 1.245676040649414, "benchmark": false}