| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 8.0, | |
| "eval_steps": 20000, | |
| "global_step": 101152, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 4.980149277433699e-05, | |
| "loss": 0.0094, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 4.9404478323010964e-05, | |
| "loss": 0.0102, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 4.900746387168493e-05, | |
| "loss": 0.0098, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 4.86104494203589e-05, | |
| "loss": 0.0101, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 4.821343496903287e-05, | |
| "loss": 0.0098, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 4.781642051770684e-05, | |
| "loss": 0.0097, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 4.741940606638082e-05, | |
| "loss": 0.0099, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 4.702239161505479e-05, | |
| "loss": 0.01, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 4.6625377163728765e-05, | |
| "loss": 0.01, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 4.6228362712402734e-05, | |
| "loss": 0.0119, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 4.5831348261076704e-05, | |
| "loss": 0.0095, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 4.543433380975068e-05, | |
| "loss": 0.0097, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "learning_rate": 4.503731935842465e-05, | |
| "loss": 0.0095, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "learning_rate": 4.464030490709862e-05, | |
| "loss": 0.0089, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 4.424329045577259e-05, | |
| "loss": 0.0088, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 4.3846276004446566e-05, | |
| "loss": 0.0088, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 4.3449261553120535e-05, | |
| "loss": 0.0085, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 4.3052247101794505e-05, | |
| "loss": 0.0087, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 4.265523265046848e-05, | |
| "loss": 0.009, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 4.225821819914245e-05, | |
| "loss": 0.0087, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_accuracy": 0.5290210343191523, | |
| "eval_exact_match": 0.5290210343191523, | |
| "eval_f1": 0.531319521324266, | |
| "eval_loss": 0.005351942032575607, | |
| "eval_precision": 0.5358216036691444, | |
| "eval_recall": 0.5290210343191523, | |
| "eval_runtime": 356.516, | |
| "eval_samples_per_second": 17.736, | |
| "eval_steps_per_second": 8.869, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 4.186120374781643e-05, | |
| "loss": 0.0082, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 4.14641892964904e-05, | |
| "loss": 0.0084, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 4.1067174845164366e-05, | |
| "loss": 0.0082, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "learning_rate": 4.0670160393838336e-05, | |
| "loss": 0.0084, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "learning_rate": 4.0273145942512306e-05, | |
| "loss": 0.0084, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "learning_rate": 3.987613149118628e-05, | |
| "loss": 0.008, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "learning_rate": 3.947911703986025e-05, | |
| "loss": 0.0075, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 3.908210258853423e-05, | |
| "loss": 0.0078, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 2.29, | |
| "learning_rate": 3.86850881372082e-05, | |
| "loss": 0.008, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 3.828807368588217e-05, | |
| "loss": 0.0075, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 3.7891059234556144e-05, | |
| "loss": 0.0077, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.53, | |
| "learning_rate": 3.749404478323011e-05, | |
| "loss": 0.0074, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 3.709703033190408e-05, | |
| "loss": 0.0078, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 3.670001588057805e-05, | |
| "loss": 0.008, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 3.630300142925203e-05, | |
| "loss": 0.0073, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 3.5905986977926e-05, | |
| "loss": 0.0076, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 3.550897252659997e-05, | |
| "loss": 0.0076, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 3.01, | |
| "learning_rate": 3.5111958075273945e-05, | |
| "loss": 0.0078, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 3.4714943623947914e-05, | |
| "loss": 0.007, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 3.431792917262189e-05, | |
| "loss": 0.0065, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "eval_accuracy": 0.5473667562865728, | |
| "eval_exact_match": 0.5473667562865728, | |
| "eval_f1": 0.5494543729242448, | |
| "eval_loss": 0.005287344101816416, | |
| "eval_precision": 0.5535347145342401, | |
| "eval_recall": 0.5473667562865728, | |
| "eval_runtime": 354.3494, | |
| "eval_samples_per_second": 17.844, | |
| "eval_steps_per_second": 8.923, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 3.24, | |
| "learning_rate": 3.3920914721295854e-05, | |
| "loss": 0.007, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 3.32, | |
| "learning_rate": 3.352390026996983e-05, | |
| "loss": 0.0071, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "learning_rate": 3.31268858186438e-05, | |
| "loss": 0.0074, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 3.48, | |
| "learning_rate": 3.272987136731777e-05, | |
| "loss": 0.0068, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 3.56, | |
| "learning_rate": 3.2332856915991746e-05, | |
| "loss": 0.0067, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 3.1935842464665715e-05, | |
| "loss": 0.0071, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 3.72, | |
| "learning_rate": 3.153882801333969e-05, | |
| "loss": 0.0067, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "learning_rate": 3.114181356201366e-05, | |
| "loss": 0.0071, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 3.88, | |
| "learning_rate": 3.074479911068763e-05, | |
| "loss": 0.0073, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 3.95, | |
| "learning_rate": 3.03477846593616e-05, | |
| "loss": 0.0067, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 4.03, | |
| "learning_rate": 2.9950770208035574e-05, | |
| "loss": 0.0067, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 4.11, | |
| "learning_rate": 2.9553755756709543e-05, | |
| "loss": 0.0064, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 4.19, | |
| "learning_rate": 2.9156741305383516e-05, | |
| "loss": 0.0062, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 4.27, | |
| "learning_rate": 2.875972685405749e-05, | |
| "loss": 0.0066, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 4.35, | |
| "learning_rate": 2.8362712402731462e-05, | |
| "loss": 0.0062, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 4.43, | |
| "learning_rate": 2.7965697951405435e-05, | |
| "loss": 0.0065, | |
| "step": 56000 | |
| }, | |
| { | |
| "epoch": 4.51, | |
| "learning_rate": 2.7568683500079405e-05, | |
| "loss": 0.0066, | |
| "step": 57000 | |
| }, | |
| { | |
| "epoch": 4.59, | |
| "learning_rate": 2.7171669048753375e-05, | |
| "loss": 0.0063, | |
| "step": 58000 | |
| }, | |
| { | |
| "epoch": 4.67, | |
| "learning_rate": 2.6774654597427344e-05, | |
| "loss": 0.0064, | |
| "step": 59000 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "learning_rate": 2.6377640146101317e-05, | |
| "loss": 0.006, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 4.75, | |
| "eval_accuracy": 0.5701407559702673, | |
| "eval_exact_match": 0.5701407559702673, | |
| "eval_f1": 0.5724919605672413, | |
| "eval_loss": 0.0050529008731245995, | |
| "eval_precision": 0.5770994780958406, | |
| "eval_recall": 0.5701407559702673, | |
| "eval_runtime": 354.5246, | |
| "eval_samples_per_second": 17.835, | |
| "eval_steps_per_second": 8.919, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 2.598062569477529e-05, | |
| "loss": 0.0065, | |
| "step": 61000 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "learning_rate": 2.5583611243449263e-05, | |
| "loss": 0.0063, | |
| "step": 62000 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 2.5186596792123236e-05, | |
| "loss": 0.0068, | |
| "step": 63000 | |
| }, | |
| { | |
| "epoch": 5.06, | |
| "learning_rate": 2.4789582340797206e-05, | |
| "loss": 0.006, | |
| "step": 64000 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 2.4392567889471176e-05, | |
| "loss": 0.006, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 5.22, | |
| "learning_rate": 2.399555343814515e-05, | |
| "loss": 0.0058, | |
| "step": 66000 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "learning_rate": 2.359853898681912e-05, | |
| "loss": 0.0056, | |
| "step": 67000 | |
| }, | |
| { | |
| "epoch": 5.38, | |
| "learning_rate": 2.3201524535493095e-05, | |
| "loss": 0.0057, | |
| "step": 68000 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 2.2804510084167068e-05, | |
| "loss": 0.0059, | |
| "step": 69000 | |
| }, | |
| { | |
| "epoch": 5.54, | |
| "learning_rate": 2.2407495632841037e-05, | |
| "loss": 0.0057, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "learning_rate": 2.2010481181515007e-05, | |
| "loss": 0.0058, | |
| "step": 71000 | |
| }, | |
| { | |
| "epoch": 5.69, | |
| "learning_rate": 2.161346673018898e-05, | |
| "loss": 0.0059, | |
| "step": 72000 | |
| }, | |
| { | |
| "epoch": 5.77, | |
| "learning_rate": 2.1216452278862953e-05, | |
| "loss": 0.0059, | |
| "step": 73000 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 2.0819437827536923e-05, | |
| "loss": 0.006, | |
| "step": 74000 | |
| }, | |
| { | |
| "epoch": 5.93, | |
| "learning_rate": 2.0422423376210896e-05, | |
| "loss": 0.0056, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 6.01, | |
| "learning_rate": 2.002540892488487e-05, | |
| "loss": 0.0058, | |
| "step": 76000 | |
| }, | |
| { | |
| "epoch": 6.09, | |
| "learning_rate": 1.9628394473558838e-05, | |
| "loss": 0.0054, | |
| "step": 77000 | |
| }, | |
| { | |
| "epoch": 6.17, | |
| "learning_rate": 1.9231380022232808e-05, | |
| "loss": 0.0061, | |
| "step": 78000 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "learning_rate": 1.883436557090678e-05, | |
| "loss": 0.0057, | |
| "step": 79000 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "learning_rate": 1.8437351119580754e-05, | |
| "loss": 0.0053, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 6.33, | |
| "eval_accuracy": 0.5827929780167642, | |
| "eval_exact_match": 0.5827929780167642, | |
| "eval_f1": 0.5851441826137382, | |
| "eval_loss": 0.004922935273498297, | |
| "eval_precision": 0.5897517001423375, | |
| "eval_recall": 0.5827929780167642, | |
| "eval_runtime": 354.2336, | |
| "eval_samples_per_second": 17.85, | |
| "eval_steps_per_second": 8.926, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 6.41, | |
| "learning_rate": 1.8040336668254727e-05, | |
| "loss": 0.0054, | |
| "step": 81000 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 1.76433222169287e-05, | |
| "loss": 0.0054, | |
| "step": 82000 | |
| }, | |
| { | |
| "epoch": 6.56, | |
| "learning_rate": 1.7246307765602666e-05, | |
| "loss": 0.0056, | |
| "step": 83000 | |
| }, | |
| { | |
| "epoch": 6.64, | |
| "learning_rate": 1.684929331427664e-05, | |
| "loss": 0.0055, | |
| "step": 84000 | |
| }, | |
| { | |
| "epoch": 6.72, | |
| "learning_rate": 1.6452278862950612e-05, | |
| "loss": 0.0051, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "learning_rate": 1.6055264411624585e-05, | |
| "loss": 0.0057, | |
| "step": 86000 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "learning_rate": 1.5658249960298555e-05, | |
| "loss": 0.0053, | |
| "step": 87000 | |
| }, | |
| { | |
| "epoch": 6.96, | |
| "learning_rate": 1.5261235508972528e-05, | |
| "loss": 0.0053, | |
| "step": 88000 | |
| }, | |
| { | |
| "epoch": 7.04, | |
| "learning_rate": 1.4864221057646499e-05, | |
| "loss": 0.0049, | |
| "step": 89000 | |
| }, | |
| { | |
| "epoch": 7.12, | |
| "learning_rate": 1.4467206606320472e-05, | |
| "loss": 0.0054, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "learning_rate": 1.4070192154994443e-05, | |
| "loss": 0.0052, | |
| "step": 91000 | |
| }, | |
| { | |
| "epoch": 7.28, | |
| "learning_rate": 1.3673177703668413e-05, | |
| "loss": 0.0052, | |
| "step": 92000 | |
| }, | |
| { | |
| "epoch": 7.36, | |
| "learning_rate": 1.3276163252342386e-05, | |
| "loss": 0.0053, | |
| "step": 93000 | |
| }, | |
| { | |
| "epoch": 7.43, | |
| "learning_rate": 1.2879148801016357e-05, | |
| "loss": 0.005, | |
| "step": 94000 | |
| }, | |
| { | |
| "epoch": 7.51, | |
| "learning_rate": 1.2482134349690329e-05, | |
| "loss": 0.0054, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 7.59, | |
| "learning_rate": 1.2085119898364302e-05, | |
| "loss": 0.0052, | |
| "step": 96000 | |
| }, | |
| { | |
| "epoch": 7.67, | |
| "learning_rate": 1.1688105447038273e-05, | |
| "loss": 0.0051, | |
| "step": 97000 | |
| }, | |
| { | |
| "epoch": 7.75, | |
| "learning_rate": 1.1291090995712244e-05, | |
| "loss": 0.005, | |
| "step": 98000 | |
| }, | |
| { | |
| "epoch": 7.83, | |
| "learning_rate": 1.0894076544386216e-05, | |
| "loss": 0.005, | |
| "step": 99000 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "learning_rate": 1.0497062093060187e-05, | |
| "loss": 0.005, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 7.91, | |
| "eval_accuracy": 0.5921239917760557, | |
| "eval_exact_match": 0.5921239917760557, | |
| "eval_f1": 0.594369761189309, | |
| "eval_loss": 0.004835059866309166, | |
| "eval_precision": 0.5987664083504666, | |
| "eval_recall": 0.5921239917760557, | |
| "eval_runtime": 354.4854, | |
| "eval_samples_per_second": 17.837, | |
| "eval_steps_per_second": 8.92, | |
| "step": 100000 | |
| }, | |
| { | |
| "epoch": 7.99, | |
| "learning_rate": 1.010004764173416e-05, | |
| "loss": 0.005, | |
| "step": 101000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 126440, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 1.0696817746378752e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |