Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -73,3 +73,9 @@
 {"step": 1460, "tokens": 765460480, "loss": 3.3235294342041017, "tokens_per_second": 979262.9726115753, "lr": 0.001095, "gradient_norm": 0.19009841978549957, "benchmark": false}
 {"step": 1480, "tokens": 775946240, "loss": 3.2929874420166017, "tokens_per_second": 986160.4285270798, "lr": 0.00111, "gradient_norm": 0.2310802936553955, "benchmark": false}
 {"step": 1500, "tokens": 786432000, "loss": 3.323834991455078, "tokens_per_second": 980533.0540742341, "lr": 0.0011250000000000001, "gradient_norm": 0.28383898735046387, "benchmark": false}
+{"step": 1520, "tokens": 796917760, "loss": 3.320008087158203, "tokens_per_second": 861185.9375980536, "lr": 0.00114, "gradient_norm": 0.28270798921585083, "benchmark": false}
+{"step": 1540, "tokens": 807403520, "loss": 3.315888214111328, "tokens_per_second": 999094.5125709875, "lr": 0.001155, "gradient_norm": 0.2359239161014557, "benchmark": false}
+{"step": 1560, "tokens": 817889280, "loss": 3.2986686706542967, "tokens_per_second": 947697.3463826232, "lr": 0.00117, "gradient_norm": 0.24340946972370148, "benchmark": false}
+{"step": 1580, "tokens": 828375040, "loss": 3.2917301177978517, "tokens_per_second": 976602.0468014935, "lr": 0.001185, "gradient_norm": 0.2536700367927551, "benchmark": false}
+{"step": 1600, "tokens": 838860800, "loss": 3.2878997802734373, "tokens_per_second": 993151.582896867, "lr": 0.0012000000000000001, "gradient_norm": 0.26822489500045776, "benchmark": false}
+{"step": 1620, "tokens": 849346560, "loss": 3.268468475341797, "tokens_per_second": 846468.1827140138, "lr": 0.0012150000000000002, "gradient_norm": 0.25949716567993164, "benchmark": false}