| { | |
| "best_metric": 1.0, | |
| "best_model_checkpoint": "DazMashaly/output/checkpoint-409", | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 1636, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.061124694376528114, | |
| "grad_norm": 1.064210295677185, | |
| "learning_rate": 5.000000000000001e-07, | |
| "loss": 6.0673, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.12224938875305623, | |
| "grad_norm": 0.982063889503479, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 6.0659, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18337408312958436, | |
| "grad_norm": 0.9887572526931763, | |
| "learning_rate": 1.5e-06, | |
| "loss": 6.069, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.24449877750611246, | |
| "grad_norm": 1.076440691947937, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 6.0588, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3056234718826406, | |
| "grad_norm": 1.1176767349243164, | |
| "learning_rate": 2.5e-06, | |
| "loss": 6.0642, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.36674816625916873, | |
| "grad_norm": 1.0234830379486084, | |
| "learning_rate": 3e-06, | |
| "loss": 6.0708, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4278728606356968, | |
| "grad_norm": 1.090944528579712, | |
| "learning_rate": 3.5e-06, | |
| "loss": 6.0339, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.4889975550122249, | |
| "grad_norm": 1.1804206371307373, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 6.0542, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5501222493887531, | |
| "grad_norm": 1.3007625341415405, | |
| "learning_rate": 4.5e-06, | |
| "loss": 6.0549, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.6112469437652812, | |
| "grad_norm": 1.2753345966339111, | |
| "learning_rate": 5e-06, | |
| "loss": 6.0535, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.6723716381418093, | |
| "grad_norm": 1.321979284286499, | |
| "learning_rate": 5.500000000000001e-06, | |
| "loss": 6.0401, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.7334963325183375, | |
| "grad_norm": 1.1843823194503784, | |
| "learning_rate": 6e-06, | |
| "loss": 6.0463, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.7946210268948656, | |
| "grad_norm": 1.2534185647964478, | |
| "learning_rate": 6.5000000000000004e-06, | |
| "loss": 6.0248, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.8557457212713936, | |
| "grad_norm": 1.2556980848312378, | |
| "learning_rate": 7e-06, | |
| "loss": 6.045, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.9168704156479217, | |
| "grad_norm": 1.263355016708374, | |
| "learning_rate": 7.500000000000001e-06, | |
| "loss": 6.034, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.9779951100244498, | |
| "grad_norm": 1.1825202703475952, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 5.9866, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_cer": 1.0, | |
| "eval_loss": 5.615485191345215, | |
| "eval_runtime": 703.3832, | |
| "eval_samples_per_second": 3.126, | |
| "eval_steps_per_second": 0.026, | |
| "eval_wer": 1.0, | |
| "step": 409 | |
| }, | |
| { | |
| "epoch": 1.039119804400978, | |
| "grad_norm": 1.388098955154419, | |
| "learning_rate": 8.5e-06, | |
| "loss": 6.0005, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.1002444987775062, | |
| "grad_norm": 1.1844673156738281, | |
| "learning_rate": 9e-06, | |
| "loss": 5.9922, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.1613691931540342, | |
| "grad_norm": 1.2709770202636719, | |
| "learning_rate": 9.5e-06, | |
| "loss": 5.9917, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.2224938875305624, | |
| "grad_norm": 1.331032395362854, | |
| "learning_rate": 1e-05, | |
| "loss": 5.9767, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.2836185819070904, | |
| "grad_norm": 1.436553955078125, | |
| "learning_rate": 9.967447916666668e-06, | |
| "loss": 5.984, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.3447432762836184, | |
| "grad_norm": 1.2195194959640503, | |
| "learning_rate": 9.934895833333335e-06, | |
| "loss": 5.9959, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.4058679706601467, | |
| "grad_norm": 1.265336036682129, | |
| "learning_rate": 9.902343750000001e-06, | |
| "loss": 5.9657, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.466992665036675, | |
| "grad_norm": 1.2491796016693115, | |
| "learning_rate": 9.869791666666667e-06, | |
| "loss": 5.9545, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.528117359413203, | |
| "grad_norm": 1.5214732885360718, | |
| "learning_rate": 9.837239583333335e-06, | |
| "loss": 5.9556, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.589242053789731, | |
| "grad_norm": 1.478154182434082, | |
| "learning_rate": 9.8046875e-06, | |
| "loss": 5.9557, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.6503667481662592, | |
| "grad_norm": 1.4325579404830933, | |
| "learning_rate": 9.772135416666667e-06, | |
| "loss": 5.9392, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.7114914425427874, | |
| "grad_norm": 1.258053183555603, | |
| "learning_rate": 9.739583333333334e-06, | |
| "loss": 5.9329, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.7726161369193154, | |
| "grad_norm": 1.439457893371582, | |
| "learning_rate": 9.707031250000001e-06, | |
| "loss": 5.9353, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 1.8337408312958434, | |
| "grad_norm": 2.1067497730255127, | |
| "learning_rate": 9.674479166666668e-06, | |
| "loss": 5.9253, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.8948655256723717, | |
| "grad_norm": 1.6002326011657715, | |
| "learning_rate": 9.641927083333335e-06, | |
| "loss": 5.8955, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 1.9559902200488999, | |
| "grad_norm": 1.249984622001648, | |
| "learning_rate": 9.609375000000001e-06, | |
| "loss": 5.8937, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_cer": 1.0, | |
| "eval_loss": 5.5135722160339355, | |
| "eval_runtime": 702.4432, | |
| "eval_samples_per_second": 3.131, | |
| "eval_steps_per_second": 0.026, | |
| "eval_wer": 1.0, | |
| "step": 818 | |
| }, | |
| { | |
| "epoch": 2.0171149144254277, | |
| "grad_norm": 1.5596375465393066, | |
| "learning_rate": 9.576822916666667e-06, | |
| "loss": 5.8783, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.078239608801956, | |
| "grad_norm": 1.5268152952194214, | |
| "learning_rate": 9.544270833333335e-06, | |
| "loss": 5.8633, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.139364303178484, | |
| "grad_norm": 1.8570572137832642, | |
| "learning_rate": 9.51171875e-06, | |
| "loss": 5.875, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.2004889975550124, | |
| "grad_norm": 1.7126903533935547, | |
| "learning_rate": 9.479166666666667e-06, | |
| "loss": 5.844, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.26161369193154, | |
| "grad_norm": 1.4625195264816284, | |
| "learning_rate": 9.446614583333334e-06, | |
| "loss": 5.8778, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 2.3227383863080684, | |
| "grad_norm": 1.4512864351272583, | |
| "learning_rate": 9.414062500000001e-06, | |
| "loss": 5.8592, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.3838630806845966, | |
| "grad_norm": 1.7913607358932495, | |
| "learning_rate": 9.381510416666668e-06, | |
| "loss": 5.8493, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 2.444987775061125, | |
| "grad_norm": 1.9679557085037231, | |
| "learning_rate": 9.348958333333335e-06, | |
| "loss": 5.8396, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.5061124694376526, | |
| "grad_norm": 1.8194330930709839, | |
| "learning_rate": 9.316406250000002e-06, | |
| "loss": 5.8435, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 2.567237163814181, | |
| "grad_norm": 1.9096548557281494, | |
| "learning_rate": 9.283854166666667e-06, | |
| "loss": 5.8206, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.628361858190709, | |
| "grad_norm": 1.6974539756774902, | |
| "learning_rate": 9.251302083333335e-06, | |
| "loss": 5.8062, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 2.689486552567237, | |
| "grad_norm": 1.7971349954605103, | |
| "learning_rate": 9.21875e-06, | |
| "loss": 5.8371, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.750611246943765, | |
| "grad_norm": 1.5566753149032593, | |
| "learning_rate": 9.186197916666667e-06, | |
| "loss": 5.8337, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 2.8117359413202934, | |
| "grad_norm": 1.7522369623184204, | |
| "learning_rate": 9.153645833333334e-06, | |
| "loss": 5.7644, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.8728606356968216, | |
| "grad_norm": 1.545310378074646, | |
| "learning_rate": 9.121093750000001e-06, | |
| "loss": 5.7901, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 2.93398533007335, | |
| "grad_norm": 1.4376916885375977, | |
| "learning_rate": 9.088541666666666e-06, | |
| "loss": 5.7892, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.9951100244498776, | |
| "grad_norm": 1.7640162706375122, | |
| "learning_rate": 9.055989583333335e-06, | |
| "loss": 5.7924, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_cer": 1.0, | |
| "eval_loss": 5.430821895599365, | |
| "eval_runtime": 955.8928, | |
| "eval_samples_per_second": 2.3, | |
| "eval_steps_per_second": 0.019, | |
| "eval_wer": 1.0, | |
| "step": 1227 | |
| }, | |
| { | |
| "epoch": 3.056234718826406, | |
| "grad_norm": 1.6651889085769653, | |
| "learning_rate": 9.0234375e-06, | |
| "loss": 5.778, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.117359413202934, | |
| "grad_norm": 2.1398913860321045, | |
| "learning_rate": 8.990885416666667e-06, | |
| "loss": 5.75, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 3.178484107579462, | |
| "grad_norm": 1.9647698402404785, | |
| "learning_rate": 8.958333333333334e-06, | |
| "loss": 5.7652, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.23960880195599, | |
| "grad_norm": 1.7572517395019531, | |
| "learning_rate": 8.92578125e-06, | |
| "loss": 5.7267, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 3.3007334963325183, | |
| "grad_norm": 2.1968727111816406, | |
| "learning_rate": 8.893229166666667e-06, | |
| "loss": 5.7432, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 3.3618581907090466, | |
| "grad_norm": 1.857980728149414, | |
| "learning_rate": 8.860677083333334e-06, | |
| "loss": 5.7513, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 3.422982885085575, | |
| "grad_norm": 1.9934026002883911, | |
| "learning_rate": 8.828125000000001e-06, | |
| "loss": 5.7348, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.4841075794621026, | |
| "grad_norm": 2.0249619483947754, | |
| "learning_rate": 8.795572916666666e-06, | |
| "loss": 5.7324, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 3.545232273838631, | |
| "grad_norm": 2.061025381088257, | |
| "learning_rate": 8.763020833333335e-06, | |
| "loss": 5.7322, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 3.606356968215159, | |
| "grad_norm": 1.7257977724075317, | |
| "learning_rate": 8.73046875e-06, | |
| "loss": 5.7322, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 3.667481662591687, | |
| "grad_norm": 1.7211699485778809, | |
| "learning_rate": 8.697916666666667e-06, | |
| "loss": 5.6969, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 3.728606356968215, | |
| "grad_norm": 2.0788753032684326, | |
| "learning_rate": 8.665364583333334e-06, | |
| "loss": 5.7322, | |
| "step": 1525 | |
| }, | |
| { | |
| "epoch": 3.7897310513447433, | |
| "grad_norm": 1.8665424585342407, | |
| "learning_rate": 8.6328125e-06, | |
| "loss": 5.7221, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 3.8508557457212715, | |
| "grad_norm": 2.3112571239471436, | |
| "learning_rate": 8.600260416666667e-06, | |
| "loss": 5.6939, | |
| "step": 1575 | |
| }, | |
| { | |
| "epoch": 3.9119804400977998, | |
| "grad_norm": 1.8274345397949219, | |
| "learning_rate": 8.567708333333334e-06, | |
| "loss": 5.6866, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 3.9731051344743276, | |
| "grad_norm": 1.6363919973373413, | |
| "learning_rate": 8.535156250000001e-06, | |
| "loss": 5.7078, | |
| "step": 1625 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_cer": 1.0, | |
| "eval_loss": 5.3543291091918945, | |
| "eval_runtime": 973.8534, | |
| "eval_samples_per_second": 2.258, | |
| "eval_steps_per_second": 0.018, | |
| "eval_wer": 1.0, | |
| "step": 1636 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 8180, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.31574988210176e+19, | |
| "train_batch_size": 124, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |