{ "final": { "optimizer": "muon", "final_val_loss": 0.8434540630146705, "final_val_ppl": 2.324381687974423, "final_test_loss": 0.850033993893169, "final_test_ppl": 2.339726386982974, "final_test_bleu": 71.09565520376408, "total_tokens": 836214, "total_steps": 3731, "trainable_parameters": 33513984 }, "interval_metrics": [ { "tokens": 83757, "step": 373, "val_loss": 11.57468023943866, "val_ppl": 106370.14659848504, "test_bleu": 56.0944193092126, "fraction": 0.1 }, { "tokens": 167424, "step": 746, "val_loss": 8.243631811956272, "val_ppl": 3803.3282283820417, "test_bleu": 57.44706269580744, "fraction": 0.2 }, { "tokens": 251309, "step": 1119, "val_loss": 3.3473492769978246, "val_ppl": 28.427280838495825, "test_bleu": 49.369356802647964, "fraction": 0.3 }, { "tokens": 335115, "step": 1492, "val_loss": 2.123773585775422, "val_ppl": 8.362635141776954, "test_bleu": 61.01711442590507, "fraction": 0.4 }, { "tokens": 418695, "step": 1866, "val_loss": 1.2813084991437111, "val_ppl": 3.6013490059431708, "test_bleu": 71.23000806985263, "fraction": 0.5 }, { "tokens": 502198, "step": 2239, "val_loss": 0.9691135202689114, "val_ppl": 2.6356070112624486, "test_bleu": 66.70788255446163, "fraction": 0.6 }, { "tokens": 585487, "step": 2612, "val_loss": 0.9083097418090074, "val_ppl": 2.480126933695958, "test_bleu": 67.52653732743481, "fraction": 0.7 }, { "tokens": 668923, "step": 2985, "val_loss": 0.8785393822485219, "val_ppl": 2.4073808739691436, "test_bleu": 73.20136104273496, "fraction": 0.8 }, { "tokens": 752480, "step": 3358, "val_loss": 0.8547548716652003, "val_ppl": 2.350798062727304, "test_bleu": 66.70788255446163, "fraction": 0.9 }, { "tokens": 836214, "step": 3731, "val_loss": 0.8434540630146705, "val_ppl": 2.324381687974423, "test_bleu": 69.07988653492116, "fraction": 1.0 } ] }