Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -23,3 +23,7 @@
 {"step": 460, "tokens": 241172480, "loss": 4.510102844238281, "tokens_per_second": 988569.6225406213, "lr": 0.00034500000000000004, "gradient_norm": 1.1452369689941406, "benchmark": false}
 {"step": 480, "tokens": 251658240, "loss": 4.4276611328125, "tokens_per_second": 989866.3254340036, "lr": 0.00035999999999999997, "gradient_norm": 0.9811086058616638, "benchmark": false}
 {"step": 500, "tokens": 262144000, "loss": 4.3114372253417965, "tokens_per_second": 987284.1551733142, "lr": 0.000375, "gradient_norm": 0.8948100805282593, "benchmark": false}
+{"step": 520, "tokens": 272629760, "loss": 4.256084060668945, "tokens_per_second": 887677.7616114634, "lr": 0.00039000000000000005, "gradient_norm": 0.8738653063774109, "benchmark": false}
+{"step": 540, "tokens": 283115520, "loss": 4.178365707397461, "tokens_per_second": 987055.5145676823, "lr": 0.00040500000000000003, "gradient_norm": 0.9119740724563599, "benchmark": false}
+{"step": 560, "tokens": 293601280, "loss": 4.115958404541016, "tokens_per_second": 986620.6540439959, "lr": 0.00042000000000000007, "gradient_norm": 0.8756295442581177, "benchmark": false}
+{"step": 580, "tokens": 304087040, "loss": 4.041319274902344, "tokens_per_second": 991743.1745662729, "lr": 0.000435, "gradient_norm": 0.7584083676338196, "benchmark": false}