Update tokenizer, configuration and training reports

reports/run/metrics.jsonl

@@ -798,3 +798,8 @@
 {"step": 15960, "tokens": 8367636480, "loss": 2.7853540420532226, "tokens_per_second": 988360.838972838, "lr": 0.0015, "gradient_norm": 0.12756744027137756, "benchmark": false}
 {"step": 15980, "tokens": 8378122240, "loss": 2.7640474319458006, "tokens_per_second": 981558.347332284, "lr": 0.0015, "gradient_norm": 0.12450078874826431, "benchmark": false}
 {"step": 16000, "tokens": 8388608000, "loss": 2.765132522583008, "tokens_per_second": 986364.2098637007, "lr": 0.0015, "gradient_norm": 0.11989699304103851, "benchmark": false}
+{"step": 16020, "tokens": 8399093760, "loss": 2.7635297775268555, "tokens_per_second": 721194.026507423, "lr": 0.0015, "gradient_norm": 0.13609609007835388, "benchmark": false}
+{"step": 16040, "tokens": 8409579520, "loss": 2.771346855163574, "tokens_per_second": 1000970.2386190228, "lr": 0.0015, "gradient_norm": 0.11046281456947327, "benchmark": false}
+{"step": 16060, "tokens": 8420065280, "loss": 2.737568664550781, "tokens_per_second": 992158.553045272, "lr": 0.0015, "gradient_norm": 0.1417708545923233, "benchmark": false}
+{"step": 16080, "tokens": 8430551040, "loss": 2.7849164962768556, "tokens_per_second": 983467.9837204237, "lr": 0.0015, "gradient_norm": 0.1339673548936844, "benchmark": false}
+{"step": 16100, "tokens": 8441036800, "loss": 2.782582092285156, "tokens_per_second": 993323.0174545496, "lr": 0.0015, "gradient_norm": 0.12592069804668427, "benchmark": false}