| { | |
| "best_metric": 0.1389758437871933, | |
| "best_model_checkpoint": "uk-mt5-base-gec-synthetic/checkpoint-12300", | |
| "epoch": 12.758942457231726, | |
| "global_step": 12300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.000979253112033195, | |
| "loss": 1.9255, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_google_bleu": 0.2703240096475649, | |
| "eval_loss": 0.9424494504928589, | |
| "eval_runtime": 347.3208, | |
| "eval_samples_per_second": 44.429, | |
| "eval_steps_per_second": 1.391, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0009585062240663901, | |
| "loss": 1.0814, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "eval_google_bleu": 0.2826040990979691, | |
| "eval_loss": 0.7668491005897522, | |
| "eval_runtime": 345.1153, | |
| "eval_samples_per_second": 44.713, | |
| "eval_steps_per_second": 1.4, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 0.000937759336099585, | |
| "loss": 0.9355, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_google_bleu": 0.2865646045540358, | |
| "eval_loss": 0.7197995185852051, | |
| "eval_runtime": 342.8829, | |
| "eval_samples_per_second": 45.004, | |
| "eval_steps_per_second": 1.409, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.00091701244813278, | |
| "loss": 0.8279, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_google_bleu": 0.292388510379304, | |
| "eval_loss": 0.6961821913719177, | |
| "eval_runtime": 342.7478, | |
| "eval_samples_per_second": 45.021, | |
| "eval_steps_per_second": 1.409, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "learning_rate": 0.0008962655601659752, | |
| "loss": 0.7798, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_google_bleu": 0.2966368793800289, | |
| "eval_loss": 0.6216537952423096, | |
| "eval_runtime": 341.7328, | |
| "eval_samples_per_second": 45.155, | |
| "eval_steps_per_second": 1.413, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 0.0008755186721991702, | |
| "loss": 0.7454, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_google_bleu": 0.2970095360433254, | |
| "eval_loss": 0.5832293629646301, | |
| "eval_runtime": 342.7772, | |
| "eval_samples_per_second": 45.018, | |
| "eval_steps_per_second": 1.409, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "learning_rate": 0.0008547717842323651, | |
| "loss": 0.6828, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.18, | |
| "eval_google_bleu": 0.30616414078516274, | |
| "eval_loss": 0.5547264814376831, | |
| "eval_runtime": 342.5678, | |
| "eval_samples_per_second": 45.045, | |
| "eval_steps_per_second": 1.41, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "learning_rate": 0.0008340248962655602, | |
| "loss": 0.6261, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.49, | |
| "eval_google_bleu": 0.30804651852418713, | |
| "eval_loss": 0.5183674693107605, | |
| "eval_runtime": 342.3924, | |
| "eval_samples_per_second": 45.068, | |
| "eval_steps_per_second": 1.411, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "learning_rate": 0.0008132780082987552, | |
| "loss": 0.6187, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "eval_google_bleu": 0.3114351154903824, | |
| "eval_loss": 0.4962822198867798, | |
| "eval_runtime": 344.2093, | |
| "eval_samples_per_second": 44.83, | |
| "eval_steps_per_second": 1.403, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "learning_rate": 0.0007925311203319502, | |
| "loss": 0.5809, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.11, | |
| "eval_google_bleu": 0.3130147936777771, | |
| "eval_loss": 0.5069273114204407, | |
| "eval_runtime": 343.9406, | |
| "eval_samples_per_second": 44.865, | |
| "eval_steps_per_second": 1.404, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "learning_rate": 0.0007717842323651453, | |
| "loss": 0.5366, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 3.42, | |
| "eval_google_bleu": 0.31680302377876657, | |
| "eval_loss": 0.46470168232917786, | |
| "eval_runtime": 343.1588, | |
| "eval_samples_per_second": 44.968, | |
| "eval_steps_per_second": 1.408, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "learning_rate": 0.0007510373443983402, | |
| "loss": 0.526, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 3.73, | |
| "eval_google_bleu": 0.3175123454942687, | |
| "eval_loss": 0.4724844694137573, | |
| "eval_runtime": 344.2418, | |
| "eval_samples_per_second": 44.826, | |
| "eval_steps_per_second": 1.403, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "learning_rate": 0.0007302904564315352, | |
| "loss": 0.5137, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 4.05, | |
| "eval_google_bleu": 0.32288427448475915, | |
| "eval_loss": 0.4414692223072052, | |
| "eval_runtime": 344.0594, | |
| "eval_samples_per_second": 44.85, | |
| "eval_steps_per_second": 1.404, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "learning_rate": 0.0007095435684647303, | |
| "loss": 0.4603, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 4.36, | |
| "eval_google_bleu": 0.3268479743473163, | |
| "eval_loss": 0.43008026480674744, | |
| "eval_runtime": 342.6092, | |
| "eval_samples_per_second": 45.04, | |
| "eval_steps_per_second": 1.41, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 0.0006887966804979254, | |
| "loss": 0.4531, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "eval_google_bleu": 0.3273518151404788, | |
| "eval_loss": 0.40794768929481506, | |
| "eval_runtime": 342.7787, | |
| "eval_samples_per_second": 45.017, | |
| "eval_steps_per_second": 1.409, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 0.0006680497925311203, | |
| "loss": 0.4555, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "eval_google_bleu": 0.32905860892065497, | |
| "eval_loss": 0.39621850848197937, | |
| "eval_runtime": 343.6219, | |
| "eval_samples_per_second": 44.907, | |
| "eval_steps_per_second": 1.406, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "learning_rate": 0.0006473029045643154, | |
| "loss": 0.392, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 5.29, | |
| "eval_google_bleu": 0.3336905264492333, | |
| "eval_loss": 0.3776128590106964, | |
| "eval_runtime": 343.634, | |
| "eval_samples_per_second": 44.905, | |
| "eval_steps_per_second": 1.406, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "learning_rate": 0.0006265560165975104, | |
| "loss": 0.402, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "eval_google_bleu": 0.3371762652161784, | |
| "eval_loss": 0.3662883937358856, | |
| "eval_runtime": 343.334, | |
| "eval_samples_per_second": 44.945, | |
| "eval_steps_per_second": 1.407, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "learning_rate": 0.0006058091286307054, | |
| "loss": 0.3979, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 5.91, | |
| "eval_google_bleu": 0.34215661656562757, | |
| "eval_loss": 0.34634825587272644, | |
| "eval_runtime": 343.5776, | |
| "eval_samples_per_second": 44.913, | |
| "eval_steps_per_second": 1.406, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "learning_rate": 0.0005850622406639005, | |
| "loss": 0.3545, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 6.22, | |
| "eval_google_bleu": 0.339056903098911, | |
| "eval_loss": 0.34026840329170227, | |
| "eval_runtime": 344.1962, | |
| "eval_samples_per_second": 44.832, | |
| "eval_steps_per_second": 1.403, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "learning_rate": 0.0005643153526970954, | |
| "loss": 0.342, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 6.53, | |
| "eval_google_bleu": 0.34660828418236983, | |
| "eval_loss": 0.3360808789730072, | |
| "eval_runtime": 344.3987, | |
| "eval_samples_per_second": 44.806, | |
| "eval_steps_per_second": 1.402, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "learning_rate": 0.0005435684647302904, | |
| "loss": 0.3462, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 6.85, | |
| "eval_google_bleu": 0.3519982269928361, | |
| "eval_loss": 0.3193075954914093, | |
| "eval_runtime": 343.4724, | |
| "eval_samples_per_second": 44.926, | |
| "eval_steps_per_second": 1.406, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 7.16, | |
| "learning_rate": 0.0005228215767634855, | |
| "loss": 0.3097, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 7.16, | |
| "eval_google_bleu": 0.35084156619797907, | |
| "eval_loss": 0.3339296281337738, | |
| "eval_runtime": 345.4283, | |
| "eval_samples_per_second": 44.672, | |
| "eval_steps_per_second": 1.398, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "learning_rate": 0.0005020746887966805, | |
| "loss": 0.2937, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 7.47, | |
| "eval_google_bleu": 0.35751958227664965, | |
| "eval_loss": 0.29495447874069214, | |
| "eval_runtime": 344.049, | |
| "eval_samples_per_second": 44.851, | |
| "eval_steps_per_second": 1.404, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "learning_rate": 0.00048132780082987554, | |
| "loss": 0.2962, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 7.78, | |
| "eval_google_bleu": 0.3584187047483623, | |
| "eval_loss": 0.27855053544044495, | |
| "eval_runtime": 344.1164, | |
| "eval_samples_per_second": 44.842, | |
| "eval_steps_per_second": 1.404, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 8.09, | |
| "learning_rate": 0.00046058091286307055, | |
| "loss": 0.2795, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 8.09, | |
| "eval_google_bleu": 0.3626892660498925, | |
| "eval_loss": 0.273299902677536, | |
| "eval_runtime": 344.7788, | |
| "eval_samples_per_second": 44.756, | |
| "eval_steps_per_second": 1.401, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "learning_rate": 0.00043983402489626555, | |
| "loss": 0.2463, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "eval_google_bleu": 0.36640212432731034, | |
| "eval_loss": 0.27249810099601746, | |
| "eval_runtime": 344.7536, | |
| "eval_samples_per_second": 44.76, | |
| "eval_steps_per_second": 1.401, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "learning_rate": 0.0004190871369294606, | |
| "loss": 0.2473, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 8.71, | |
| "eval_google_bleu": 0.3680926990956975, | |
| "eval_loss": 0.25599655508995056, | |
| "eval_runtime": 343.8059, | |
| "eval_samples_per_second": 44.883, | |
| "eval_steps_per_second": 1.405, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "learning_rate": 0.00039834024896265557, | |
| "loss": 0.2462, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 9.02, | |
| "eval_google_bleu": 0.3726668030252467, | |
| "eval_loss": 0.24074989557266235, | |
| "eval_runtime": 343.4854, | |
| "eval_samples_per_second": 44.925, | |
| "eval_steps_per_second": 1.406, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 9.34, | |
| "learning_rate": 0.00037759336099585063, | |
| "loss": 0.2003, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 9.34, | |
| "eval_google_bleu": 0.37575691510452164, | |
| "eval_loss": 0.23459354043006897, | |
| "eval_runtime": 343.9506, | |
| "eval_samples_per_second": 44.864, | |
| "eval_steps_per_second": 1.404, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 9.65, | |
| "learning_rate": 0.00035684647302904564, | |
| "loss": 0.2096, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 9.65, | |
| "eval_google_bleu": 0.3810405269728534, | |
| "eval_loss": 0.23272807896137238, | |
| "eval_runtime": 345.3533, | |
| "eval_samples_per_second": 44.682, | |
| "eval_steps_per_second": 1.399, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 9.96, | |
| "learning_rate": 0.0003360995850622407, | |
| "loss": 0.2109, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 9.96, | |
| "eval_google_bleu": 0.3860127254969551, | |
| "eval_loss": 0.2076478898525238, | |
| "eval_runtime": 345.3314, | |
| "eval_samples_per_second": 44.685, | |
| "eval_steps_per_second": 1.399, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 10.27, | |
| "learning_rate": 0.00031535269709543565, | |
| "loss": 0.1673, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 10.27, | |
| "eval_google_bleu": 0.39099070617947457, | |
| "eval_loss": 0.2026001662015915, | |
| "eval_runtime": 344.3009, | |
| "eval_samples_per_second": 44.818, | |
| "eval_steps_per_second": 1.403, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 10.58, | |
| "learning_rate": 0.0002946058091286307, | |
| "loss": 0.1722, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 10.58, | |
| "eval_google_bleu": 0.39423925473127097, | |
| "eval_loss": 0.19409596920013428, | |
| "eval_runtime": 345.022, | |
| "eval_samples_per_second": 44.725, | |
| "eval_steps_per_second": 1.4, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 10.89, | |
| "learning_rate": 0.0002738589211618258, | |
| "loss": 0.1692, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 10.89, | |
| "eval_google_bleu": 0.39808409620231944, | |
| "eval_loss": 0.18527193367481232, | |
| "eval_runtime": 344.9054, | |
| "eval_samples_per_second": 44.74, | |
| "eval_steps_per_second": 1.4, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "learning_rate": 0.00025311203319502073, | |
| "loss": 0.1414, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "eval_google_bleu": 0.3988704921153549, | |
| "eval_loss": 0.18296599388122559, | |
| "eval_runtime": 344.3914, | |
| "eval_samples_per_second": 44.807, | |
| "eval_steps_per_second": 1.402, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "learning_rate": 0.00023236514522821577, | |
| "loss": 0.1357, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 11.51, | |
| "eval_google_bleu": 0.40408064479411665, | |
| "eval_loss": 0.17169393599033356, | |
| "eval_runtime": 345.2605, | |
| "eval_samples_per_second": 44.694, | |
| "eval_steps_per_second": 1.399, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 11.83, | |
| "learning_rate": 0.0002116182572614108, | |
| "loss": 0.1335, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 11.83, | |
| "eval_google_bleu": 0.4105103547726813, | |
| "eval_loss": 0.15817226469516754, | |
| "eval_runtime": 345.0386, | |
| "eval_samples_per_second": 44.723, | |
| "eval_steps_per_second": 1.4, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 12.14, | |
| "learning_rate": 0.0001908713692946058, | |
| "loss": 0.119, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 12.14, | |
| "eval_google_bleu": 0.4141623418407881, | |
| "eval_loss": 0.1521884799003601, | |
| "eval_runtime": 344.7777, | |
| "eval_samples_per_second": 44.756, | |
| "eval_steps_per_second": 1.401, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 12.45, | |
| "learning_rate": 0.00017012448132780084, | |
| "loss": 0.102, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 12.45, | |
| "eval_google_bleu": 0.4170916396229331, | |
| "eval_loss": 0.150096133351326, | |
| "eval_runtime": 344.5413, | |
| "eval_samples_per_second": 44.787, | |
| "eval_steps_per_second": 1.402, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 12.76, | |
| "learning_rate": 0.00014937759336099585, | |
| "loss": 0.1014, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 12.76, | |
| "eval_google_bleu": 0.4215356451396477, | |
| "eval_loss": 0.1389758437871933, | |
| "eval_runtime": 343.8578, | |
| "eval_samples_per_second": 44.876, | |
| "eval_steps_per_second": 1.405, | |
| "step": 12300 | |
| } | |
| ], | |
| "max_steps": 14460, | |
| "num_train_epochs": 15, | |
| "total_flos": 2.3750619195838464e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |