| { | |
| "best_global_step": 1500, | |
| "best_metric": 0.4347842335700989, | |
| "best_model_checkpoint": "./speecht5_finetuned_Darija/checkpoint-1500", | |
| "epoch": 4.178583885594699, | |
| "eval_steps": 100, | |
| "global_step": 1500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06975933031042902, | |
| "grad_norm": 4.014983654022217, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.2037, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.13951866062085805, | |
| "grad_norm": 4.588967800140381, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.8946, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.20927799093128707, | |
| "grad_norm": 4.026744842529297, | |
| "learning_rate": 7.3e-05, | |
| "loss": 0.6685, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.2790373212417161, | |
| "grad_norm": 3.659444808959961, | |
| "learning_rate": 9.8e-05, | |
| "loss": 0.6168, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.2790373212417161, | |
| "eval_loss": 0.5272690057754517, | |
| "eval_runtime": 41.6965, | |
| "eval_samples_per_second": 30.578, | |
| "eval_steps_per_second": 15.301, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.3487966515521451, | |
| "grad_norm": 6.453115940093994, | |
| "learning_rate": 9.878947368421053e-05, | |
| "loss": 0.5847, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.41855598186257414, | |
| "grad_norm": 3.9114222526550293, | |
| "learning_rate": 9.747368421052632e-05, | |
| "loss": 0.5667, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.4883153121730031, | |
| "grad_norm": 5.056523323059082, | |
| "learning_rate": 9.615789473684212e-05, | |
| "loss": 0.5662, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.5580746424834322, | |
| "grad_norm": 5.494399070739746, | |
| "learning_rate": 9.48421052631579e-05, | |
| "loss": 0.5699, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.5580746424834322, | |
| "eval_loss": 0.5231854915618896, | |
| "eval_runtime": 37.0744, | |
| "eval_samples_per_second": 34.39, | |
| "eval_steps_per_second": 17.209, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6278339727938612, | |
| "grad_norm": 2.748530864715576, | |
| "learning_rate": 9.352631578947368e-05, | |
| "loss": 0.5448, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.6975933031042902, | |
| "grad_norm": 3.9691007137298584, | |
| "learning_rate": 9.221052631578948e-05, | |
| "loss": 0.5358, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.7673526334147193, | |
| "grad_norm": 3.5358879566192627, | |
| "learning_rate": 9.089473684210526e-05, | |
| "loss": 0.5245, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.8371119637251483, | |
| "grad_norm": 2.21895432472229, | |
| "learning_rate": 8.957894736842106e-05, | |
| "loss": 0.5231, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.8371119637251483, | |
| "eval_loss": 0.47762706875801086, | |
| "eval_runtime": 36.3582, | |
| "eval_samples_per_second": 35.068, | |
| "eval_steps_per_second": 17.548, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9068712940355773, | |
| "grad_norm": 3.315195322036743, | |
| "learning_rate": 8.826315789473684e-05, | |
| "loss": 0.5349, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.9766306243460062, | |
| "grad_norm": 2.936798572540283, | |
| "learning_rate": 8.694736842105264e-05, | |
| "loss": 0.5248, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 1.0446459713986747, | |
| "grad_norm": 2.950599431991577, | |
| "learning_rate": 8.563157894736843e-05, | |
| "loss": 0.5101, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 1.1144053017091036, | |
| "grad_norm": 4.972070693969727, | |
| "learning_rate": 8.431578947368422e-05, | |
| "loss": 0.515, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.1144053017091036, | |
| "eval_loss": 0.49294987320899963, | |
| "eval_runtime": 35.8818, | |
| "eval_samples_per_second": 35.533, | |
| "eval_steps_per_second": 17.781, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 1.1841646320195327, | |
| "grad_norm": 2.6673648357391357, | |
| "learning_rate": 8.3e-05, | |
| "loss": 0.5115, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 1.2539239623299616, | |
| "grad_norm": 3.388873338699341, | |
| "learning_rate": 8.16842105263158e-05, | |
| "loss": 0.5064, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 1.3236832926403905, | |
| "grad_norm": 2.4961979389190674, | |
| "learning_rate": 8.036842105263158e-05, | |
| "loss": 0.5028, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 1.3934426229508197, | |
| "grad_norm": 2.7970707416534424, | |
| "learning_rate": 7.905263157894737e-05, | |
| "loss": 0.4975, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.3934426229508197, | |
| "eval_loss": 0.4633351266384125, | |
| "eval_runtime": 36.5245, | |
| "eval_samples_per_second": 34.908, | |
| "eval_steps_per_second": 17.468, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.4632019532612488, | |
| "grad_norm": 2.774756908416748, | |
| "learning_rate": 7.773684210526317e-05, | |
| "loss": 0.498, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 1.5329612835716777, | |
| "grad_norm": 5.114898204803467, | |
| "learning_rate": 7.642105263157895e-05, | |
| "loss": 0.4974, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.6027206138821066, | |
| "grad_norm": 3.1757712364196777, | |
| "learning_rate": 7.510526315789475e-05, | |
| "loss": 0.5068, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 1.6724799441925358, | |
| "grad_norm": 2.1318249702453613, | |
| "learning_rate": 7.378947368421053e-05, | |
| "loss": 0.5003, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.6724799441925358, | |
| "eval_loss": 0.45842912793159485, | |
| "eval_runtime": 36.8677, | |
| "eval_samples_per_second": 34.583, | |
| "eval_steps_per_second": 17.305, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.742239274502965, | |
| "grad_norm": 3.224851131439209, | |
| "learning_rate": 7.247368421052631e-05, | |
| "loss": 0.4954, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 1.8119986048133938, | |
| "grad_norm": 2.2048707008361816, | |
| "learning_rate": 7.115789473684211e-05, | |
| "loss": 0.4954, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.8817579351238227, | |
| "grad_norm": 3.310940742492676, | |
| "learning_rate": 6.98421052631579e-05, | |
| "loss": 0.4898, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 1.9515172654342519, | |
| "grad_norm": 3.6913022994995117, | |
| "learning_rate": 6.852631578947369e-05, | |
| "loss": 0.4881, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.9515172654342519, | |
| "eval_loss": 0.4606294631958008, | |
| "eval_runtime": 36.611, | |
| "eval_samples_per_second": 34.826, | |
| "eval_steps_per_second": 17.426, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 2.01953261248692, | |
| "grad_norm": 3.2030959129333496, | |
| "learning_rate": 6.721052631578948e-05, | |
| "loss": 0.4727, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 2.0892919427973493, | |
| "grad_norm": 3.458202362060547, | |
| "learning_rate": 6.589473684210526e-05, | |
| "loss": 0.4809, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 2.159051273107778, | |
| "grad_norm": 1.9468854665756226, | |
| "learning_rate": 6.457894736842106e-05, | |
| "loss": 0.4749, | |
| "step": 775 | |
| }, | |
| { | |
| "epoch": 2.228810603418207, | |
| "grad_norm": 3.895932912826538, | |
| "learning_rate": 6.331578947368422e-05, | |
| "loss": 0.4803, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.228810603418207, | |
| "eval_loss": 0.4526459872722626, | |
| "eval_runtime": 37.4188, | |
| "eval_samples_per_second": 34.074, | |
| "eval_steps_per_second": 17.05, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 2.2985699337286363, | |
| "grad_norm": 3.0414257049560547, | |
| "learning_rate": 6.2e-05, | |
| "loss": 0.4774, | |
| "step": 825 | |
| }, | |
| { | |
| "epoch": 2.3683292640390654, | |
| "grad_norm": 2.6127874851226807, | |
| "learning_rate": 6.0684210526315785e-05, | |
| "loss": 0.4745, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 2.438088594349494, | |
| "grad_norm": 3.0023088455200195, | |
| "learning_rate": 5.936842105263158e-05, | |
| "loss": 0.4791, | |
| "step": 875 | |
| }, | |
| { | |
| "epoch": 2.5078479246599232, | |
| "grad_norm": 2.730001926422119, | |
| "learning_rate": 5.805263157894737e-05, | |
| "loss": 0.4785, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.5078479246599232, | |
| "eval_loss": 0.448452353477478, | |
| "eval_runtime": 37.2061, | |
| "eval_samples_per_second": 34.269, | |
| "eval_steps_per_second": 17.148, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 2.5776072549703524, | |
| "grad_norm": 2.2317161560058594, | |
| "learning_rate": 5.6736842105263166e-05, | |
| "loss": 0.4732, | |
| "step": 925 | |
| }, | |
| { | |
| "epoch": 2.647366585280781, | |
| "grad_norm": 3.1052346229553223, | |
| "learning_rate": 5.542105263157895e-05, | |
| "loss": 0.4801, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 2.71712591559121, | |
| "grad_norm": 2.309736967086792, | |
| "learning_rate": 5.410526315789474e-05, | |
| "loss": 0.4789, | |
| "step": 975 | |
| }, | |
| { | |
| "epoch": 2.7868852459016393, | |
| "grad_norm": 2.5499184131622314, | |
| "learning_rate": 5.2789473684210534e-05, | |
| "loss": 0.4733, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.7868852459016393, | |
| "eval_loss": 0.44482412934303284, | |
| "eval_runtime": 36.9591, | |
| "eval_samples_per_second": 34.498, | |
| "eval_steps_per_second": 17.262, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.8566445762120685, | |
| "grad_norm": 3.4246134757995605, | |
| "learning_rate": 5.1473684210526317e-05, | |
| "loss": 0.4778, | |
| "step": 1025 | |
| }, | |
| { | |
| "epoch": 2.9264039065224976, | |
| "grad_norm": 2.3219287395477295, | |
| "learning_rate": 5.0157894736842106e-05, | |
| "loss": 0.4748, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 2.9961632368329263, | |
| "grad_norm": 4.338850021362305, | |
| "learning_rate": 4.8842105263157895e-05, | |
| "loss": 0.4738, | |
| "step": 1075 | |
| }, | |
| { | |
| "epoch": 3.0641785838855946, | |
| "grad_norm": 2.8604490756988525, | |
| "learning_rate": 4.7526315789473684e-05, | |
| "loss": 0.4542, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.0641785838855946, | |
| "eval_loss": 0.4392930865287781, | |
| "eval_runtime": 37.1499, | |
| "eval_samples_per_second": 34.32, | |
| "eval_steps_per_second": 17.174, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 3.1339379141960237, | |
| "grad_norm": 3.0904340744018555, | |
| "learning_rate": 4.6210526315789473e-05, | |
| "loss": 0.4644, | |
| "step": 1125 | |
| }, | |
| { | |
| "epoch": 3.203697244506453, | |
| "grad_norm": 3.41607403755188, | |
| "learning_rate": 4.489473684210527e-05, | |
| "loss": 0.4678, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 3.2734565748168816, | |
| "grad_norm": 2.6169519424438477, | |
| "learning_rate": 4.357894736842105e-05, | |
| "loss": 0.4682, | |
| "step": 1175 | |
| }, | |
| { | |
| "epoch": 3.3432159051273107, | |
| "grad_norm": 3.6565799713134766, | |
| "learning_rate": 4.226315789473684e-05, | |
| "loss": 0.4607, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.3432159051273107, | |
| "eval_loss": 0.43838903307914734, | |
| "eval_runtime": 37.1276, | |
| "eval_samples_per_second": 34.341, | |
| "eval_steps_per_second": 17.184, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 3.41297523543774, | |
| "grad_norm": 2.939362049102783, | |
| "learning_rate": 4.094736842105264e-05, | |
| "loss": 0.4586, | |
| "step": 1225 | |
| }, | |
| { | |
| "epoch": 3.482734565748169, | |
| "grad_norm": 3.203904390335083, | |
| "learning_rate": 3.9631578947368426e-05, | |
| "loss": 0.46, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 3.552493896058598, | |
| "grad_norm": 3.043788433074951, | |
| "learning_rate": 3.831578947368421e-05, | |
| "loss": 0.4544, | |
| "step": 1275 | |
| }, | |
| { | |
| "epoch": 3.622253226369027, | |
| "grad_norm": 4.401975631713867, | |
| "learning_rate": 3.7e-05, | |
| "loss": 0.4569, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.622253226369027, | |
| "eval_loss": 0.4503220319747925, | |
| "eval_runtime": 37.3024, | |
| "eval_samples_per_second": 34.18, | |
| "eval_steps_per_second": 17.103, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 3.692012556679456, | |
| "grad_norm": 2.873894453048706, | |
| "learning_rate": 3.5684210526315794e-05, | |
| "loss": 0.4582, | |
| "step": 1325 | |
| }, | |
| { | |
| "epoch": 3.7617718869898846, | |
| "grad_norm": 2.5376503467559814, | |
| "learning_rate": 3.436842105263158e-05, | |
| "loss": 0.4611, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 3.8315312173003138, | |
| "grad_norm": 2.759460687637329, | |
| "learning_rate": 3.3052631578947366e-05, | |
| "loss": 0.4649, | |
| "step": 1375 | |
| }, | |
| { | |
| "epoch": 3.901290547610743, | |
| "grad_norm": 2.2127342224121094, | |
| "learning_rate": 3.173684210526316e-05, | |
| "loss": 0.4582, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.901290547610743, | |
| "eval_loss": 0.43552732467651367, | |
| "eval_runtime": 37.4108, | |
| "eval_samples_per_second": 34.081, | |
| "eval_steps_per_second": 17.054, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 3.971049877921172, | |
| "grad_norm": 3.02252459526062, | |
| "learning_rate": 3.042105263157895e-05, | |
| "loss": 0.4639, | |
| "step": 1425 | |
| }, | |
| { | |
| "epoch": 4.03906522497384, | |
| "grad_norm": 3.38741397857666, | |
| "learning_rate": 2.910526315789474e-05, | |
| "loss": 0.4457, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 4.1088245552842695, | |
| "grad_norm": 2.798316717147827, | |
| "learning_rate": 2.7789473684210526e-05, | |
| "loss": 0.4571, | |
| "step": 1475 | |
| }, | |
| { | |
| "epoch": 4.178583885594699, | |
| "grad_norm": 2.803051471710205, | |
| "learning_rate": 2.647368421052632e-05, | |
| "loss": 0.451, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 4.178583885594699, | |
| "eval_loss": 0.4347842335700989, | |
| "eval_runtime": 37.2652, | |
| "eval_samples_per_second": 34.214, | |
| "eval_steps_per_second": 17.121, | |
| "step": 1500 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 2000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 6, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1844132330585952.0, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |