| { | |
| "best_global_step": 300, | |
| "best_metric": 2.4513747574653353, | |
| "best_model_checkpoint": "./SALAMA_NEW99/checkpoint-300", | |
| "epoch": 0.9787928221859706, | |
| "eval_steps": 300, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03262642740619902, | |
| "grad_norm": 4.136127471923828, | |
| "learning_rate": 1.8e-07, | |
| "loss": 0.0657, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.06525285481239804, | |
| "grad_norm": 3.8954381942749023, | |
| "learning_rate": 3.8e-07, | |
| "loss": 0.0652, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.09787928221859707, | |
| "grad_norm": 3.1246440410614014, | |
| "learning_rate": 5.800000000000001e-07, | |
| "loss": 0.0435, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.13050570962479607, | |
| "grad_norm": 5.918012619018555, | |
| "learning_rate": 7.8e-07, | |
| "loss": 0.071, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.1631321370309951, | |
| "grad_norm": 3.991386890411377, | |
| "learning_rate": 9.800000000000001e-07, | |
| "loss": 0.0554, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.19575856443719414, | |
| "grad_norm": 3.4036903381347656, | |
| "learning_rate": 1.1800000000000001e-06, | |
| "loss": 0.0729, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.22838499184339314, | |
| "grad_norm": 3.296243906021118, | |
| "learning_rate": 1.3800000000000001e-06, | |
| "loss": 0.0683, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.26101141924959215, | |
| "grad_norm": 4.985351085662842, | |
| "learning_rate": 1.5800000000000001e-06, | |
| "loss": 0.0635, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.2936378466557912, | |
| "grad_norm": 5.250389575958252, | |
| "learning_rate": 1.7800000000000001e-06, | |
| "loss": 0.0864, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.3262642740619902, | |
| "grad_norm": 3.3003408908843994, | |
| "learning_rate": 1.98e-06, | |
| "loss": 0.0582, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.35889070146818924, | |
| "grad_norm": 5.999060153961182, | |
| "learning_rate": 2.1800000000000003e-06, | |
| "loss": 0.073, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.3915171288743883, | |
| "grad_norm": 2.205695152282715, | |
| "learning_rate": 2.38e-06, | |
| "loss": 0.055, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.42414355628058725, | |
| "grad_norm": 3.037153482437134, | |
| "learning_rate": 2.5800000000000003e-06, | |
| "loss": 0.0541, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.4567699836867863, | |
| "grad_norm": 5.886940002441406, | |
| "learning_rate": 2.7800000000000005e-06, | |
| "loss": 0.0593, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.4893964110929853, | |
| "grad_norm": 3.1995198726654053, | |
| "learning_rate": 2.9800000000000003e-06, | |
| "loss": 0.0566, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.5220228384991843, | |
| "grad_norm": 1.9580050706863403, | |
| "learning_rate": 3.1800000000000005e-06, | |
| "loss": 0.0533, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.5546492659053833, | |
| "grad_norm": 3.5993852615356445, | |
| "learning_rate": 3.3800000000000007e-06, | |
| "loss": 0.0723, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.5872756933115824, | |
| "grad_norm": 4.075970649719238, | |
| "learning_rate": 3.58e-06, | |
| "loss": 0.0671, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.6199021207177814, | |
| "grad_norm": 5.4311347007751465, | |
| "learning_rate": 3.7800000000000002e-06, | |
| "loss": 0.0751, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.6525285481239804, | |
| "grad_norm": 4.403566837310791, | |
| "learning_rate": 3.980000000000001e-06, | |
| "loss": 0.0571, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.6851549755301795, | |
| "grad_norm": 3.001912832260132, | |
| "learning_rate": 4.18e-06, | |
| "loss": 0.0697, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.7177814029363785, | |
| "grad_norm": 6.647680759429932, | |
| "learning_rate": 4.38e-06, | |
| "loss": 0.0636, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.7504078303425775, | |
| "grad_norm": 2.4966087341308594, | |
| "learning_rate": 4.58e-06, | |
| "loss": 0.0524, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.7830342577487766, | |
| "grad_norm": 5.768764019012451, | |
| "learning_rate": 4.78e-06, | |
| "loss": 0.042, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.8156606851549756, | |
| "grad_norm": 4.395790100097656, | |
| "learning_rate": 4.980000000000001e-06, | |
| "loss": 0.0703, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.8482871125611745, | |
| "grad_norm": 3.6472349166870117, | |
| "learning_rate": 5.18e-06, | |
| "loss": 0.055, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.8809135399673735, | |
| "grad_norm": 4.439610958099365, | |
| "learning_rate": 5.380000000000001e-06, | |
| "loss": 0.0719, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.9135399673735726, | |
| "grad_norm": 3.3142175674438477, | |
| "learning_rate": 5.580000000000001e-06, | |
| "loss": 0.0587, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.9461663947797716, | |
| "grad_norm": 2.1484858989715576, | |
| "learning_rate": 5.78e-06, | |
| "loss": 0.0491, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.9787928221859706, | |
| "grad_norm": 3.151224136352539, | |
| "learning_rate": 5.98e-06, | |
| "loss": 0.0597, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.9787928221859706, | |
| "eval_loss": 0.035220563411712646, | |
| "eval_runtime": 877.1298, | |
| "eval_samples_per_second": 2.794, | |
| "eval_steps_per_second": 0.35, | |
| "eval_wer": 2.4513747574653353, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 921, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 300, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.770419843072e+18, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |