| { | |
| "best_metric": 0.4299218952655792, | |
| "best_model_checkpoint": "/mnt/beegfs/farid/mlora/outputs/xnli/aya-101/fr/rank4_lr5e-5/checkpoint-6000", | |
| "epoch": 0.24445893089960888, | |
| "eval_steps": 500, | |
| "global_step": 6000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.020371577574967405, | |
| "grad_norm": 4.124022483825684, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.1088, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.020371577574967405, | |
| "eval_accuracy": 0.6236947791164659, | |
| "eval_f1": 0.5758401803622193, | |
| "eval_loss": 0.8937909603118896, | |
| "eval_runtime": 414.8928, | |
| "eval_samples_per_second": 6.002, | |
| "eval_steps_per_second": 0.376, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.04074315514993481, | |
| "grad_norm": 11.531847953796387, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.6465, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04074315514993481, | |
| "eval_accuracy": 0.7281124497991968, | |
| "eval_f1": 0.7017662443726991, | |
| "eval_loss": 0.6295766234397888, | |
| "eval_runtime": 411.581, | |
| "eval_samples_per_second": 6.05, | |
| "eval_steps_per_second": 0.379, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.06111473272490222, | |
| "grad_norm": 20.772192001342773, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.5523, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.06111473272490222, | |
| "eval_accuracy": 0.8004016064257028, | |
| "eval_f1": 0.7942881753796405, | |
| "eval_loss": 0.5290858745574951, | |
| "eval_runtime": 411.8482, | |
| "eval_samples_per_second": 6.046, | |
| "eval_steps_per_second": 0.379, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.08148631029986962, | |
| "grad_norm": 9.07897663116455, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.5112, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.08148631029986962, | |
| "eval_accuracy": 0.8164658634538152, | |
| "eval_f1": 0.8113021982022444, | |
| "eval_loss": 0.5029298067092896, | |
| "eval_runtime": 431.4155, | |
| "eval_samples_per_second": 5.772, | |
| "eval_steps_per_second": 0.362, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.10185788787483703, | |
| "grad_norm": 12.577387809753418, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.4985, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.10185788787483703, | |
| "eval_accuracy": 0.8160642570281125, | |
| "eval_f1": 0.8112429470654366, | |
| "eval_loss": 0.509430468082428, | |
| "eval_runtime": 411.5028, | |
| "eval_samples_per_second": 6.051, | |
| "eval_steps_per_second": 0.379, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.12222946544980444, | |
| "grad_norm": 14.661023139953613, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.4781, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.12222946544980444, | |
| "eval_accuracy": 0.8349397590361446, | |
| "eval_f1": 0.8324700562884065, | |
| "eval_loss": 0.4642679691314697, | |
| "eval_runtime": 411.3144, | |
| "eval_samples_per_second": 6.054, | |
| "eval_steps_per_second": 0.379, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.14260104302477183, | |
| "grad_norm": 10.473104476928711, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.4876, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.14260104302477183, | |
| "eval_accuracy": 0.8369477911646587, | |
| "eval_f1": 0.8347534979362655, | |
| "eval_loss": 0.4561132788658142, | |
| "eval_runtime": 411.1784, | |
| "eval_samples_per_second": 6.056, | |
| "eval_steps_per_second": 0.379, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.16297262059973924, | |
| "grad_norm": 6.872061729431152, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.4851, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.16297262059973924, | |
| "eval_accuracy": 0.8429718875502008, | |
| "eval_f1": 0.8409600083623853, | |
| "eval_loss": 0.44605061411857605, | |
| "eval_runtime": 411.3811, | |
| "eval_samples_per_second": 6.053, | |
| "eval_steps_per_second": 0.379, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.18334419817470665, | |
| "grad_norm": 10.347468376159668, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.4647, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.18334419817470665, | |
| "eval_accuracy": 0.8449799196787149, | |
| "eval_f1": 0.843138936718116, | |
| "eval_loss": 0.4416460692882538, | |
| "eval_runtime": 411.5002, | |
| "eval_samples_per_second": 6.051, | |
| "eval_steps_per_second": 0.379, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.20371577574967406, | |
| "grad_norm": 5.183737277984619, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.4771, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.20371577574967406, | |
| "eval_accuracy": 0.8409638554216867, | |
| "eval_f1": 0.8392109978727667, | |
| "eval_loss": 0.4357330799102783, | |
| "eval_runtime": 411.6006, | |
| "eval_samples_per_second": 6.05, | |
| "eval_steps_per_second": 0.379, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.22408735332464147, | |
| "grad_norm": 6.372146129608154, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.4784, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.22408735332464147, | |
| "eval_accuracy": 0.8477911646586346, | |
| "eval_f1": 0.846239279407909, | |
| "eval_loss": 0.43362346291542053, | |
| "eval_runtime": 411.218, | |
| "eval_samples_per_second": 6.055, | |
| "eval_steps_per_second": 0.379, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.24445893089960888, | |
| "grad_norm": 8.003700256347656, | |
| "learning_rate": 0.0, | |
| "loss": 0.4498, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.24445893089960888, | |
| "eval_accuracy": 0.8465863453815261, | |
| "eval_f1": 0.8452045185120488, | |
| "eval_loss": 0.4299218952655792, | |
| "eval_runtime": 411.2051, | |
| "eval_samples_per_second": 6.055, | |
| "eval_steps_per_second": 0.379, | |
| "step": 6000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 6000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "total_flos": 8.03166870528e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |