| { | |
| "best_metric": 0.5128149390220642, | |
| "best_model_checkpoint": "/mnt/beegfs/farid/mlora/outputs/xnli/aya-101/hi/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": 1.9568703174591064, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.1534, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.020371577574967405, | |
| "eval_accuracy": 0.37991967871485943, | |
| "eval_f1": 0.3396581059481989, | |
| "eval_loss": 1.0917378664016724, | |
| "eval_runtime": 413.6608, | |
| "eval_samples_per_second": 6.019, | |
| "eval_steps_per_second": 0.377, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.04074315514993481, | |
| "grad_norm": 9.426871299743652, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 1.0218, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04074315514993481, | |
| "eval_accuracy": 0.6409638554216868, | |
| "eval_f1": 0.6436728838823977, | |
| "eval_loss": 0.8600370287895203, | |
| "eval_runtime": 412.0934, | |
| "eval_samples_per_second": 6.042, | |
| "eval_steps_per_second": 0.379, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.06111473272490222, | |
| "grad_norm": 20.413841247558594, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.8582, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.06111473272490222, | |
| "eval_accuracy": 0.7369477911646586, | |
| "eval_f1": 0.7370571656251429, | |
| "eval_loss": 0.6784067749977112, | |
| "eval_runtime": 431.3062, | |
| "eval_samples_per_second": 5.773, | |
| "eval_steps_per_second": 0.362, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.08148631029986962, | |
| "grad_norm": 8.49395751953125, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.7793, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.08148631029986962, | |
| "eval_accuracy": 0.7682730923694779, | |
| "eval_f1": 0.7693110628872669, | |
| "eval_loss": 0.611162006855011, | |
| "eval_runtime": 431.4602, | |
| "eval_samples_per_second": 5.771, | |
| "eval_steps_per_second": 0.362, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.10185788787483703, | |
| "grad_norm": 7.8166375160217285, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.7563, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.10185788787483703, | |
| "eval_accuracy": 0.7871485943775101, | |
| "eval_f1": 0.7881188800855449, | |
| "eval_loss": 0.5777685046195984, | |
| "eval_runtime": 411.445, | |
| "eval_samples_per_second": 6.052, | |
| "eval_steps_per_second": 0.379, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.12222946544980444, | |
| "grad_norm": 11.658799171447754, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.7191, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.12222946544980444, | |
| "eval_accuracy": 0.7971887550200804, | |
| "eval_f1": 0.7971744387340783, | |
| "eval_loss": 0.5552772283554077, | |
| "eval_runtime": 411.1455, | |
| "eval_samples_per_second": 6.056, | |
| "eval_steps_per_second": 0.379, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.14260104302477183, | |
| "grad_norm": 8.530998229980469, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.7052, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.14260104302477183, | |
| "eval_accuracy": 0.8016064257028113, | |
| "eval_f1": 0.8021556716106938, | |
| "eval_loss": 0.5359864234924316, | |
| "eval_runtime": 411.8163, | |
| "eval_samples_per_second": 6.046, | |
| "eval_steps_per_second": 0.379, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.16297262059973924, | |
| "grad_norm": 6.73158597946167, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.7079, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.16297262059973924, | |
| "eval_accuracy": 0.802008032128514, | |
| "eval_f1": 0.8023252480910511, | |
| "eval_loss": 0.5278254747390747, | |
| "eval_runtime": 411.3361, | |
| "eval_samples_per_second": 6.053, | |
| "eval_steps_per_second": 0.379, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.18334419817470665, | |
| "grad_norm": 9.069704055786133, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.6775, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.18334419817470665, | |
| "eval_accuracy": 0.8036144578313253, | |
| "eval_f1": 0.8043620274591458, | |
| "eval_loss": 0.5234741568565369, | |
| "eval_runtime": 431.5576, | |
| "eval_samples_per_second": 5.77, | |
| "eval_steps_per_second": 0.361, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.20371577574967406, | |
| "grad_norm": 9.322051048278809, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.6977, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.20371577574967406, | |
| "eval_accuracy": 0.8088353413654619, | |
| "eval_f1": 0.8089545270025741, | |
| "eval_loss": 0.5214890837669373, | |
| "eval_runtime": 412.4688, | |
| "eval_samples_per_second": 6.037, | |
| "eval_steps_per_second": 0.378, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.22408735332464147, | |
| "grad_norm": 5.2376508712768555, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.6834, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.22408735332464147, | |
| "eval_accuracy": 0.8076305220883534, | |
| "eval_f1": 0.8082794716875278, | |
| "eval_loss": 0.5130343437194824, | |
| "eval_runtime": 411.1989, | |
| "eval_samples_per_second": 6.055, | |
| "eval_steps_per_second": 0.379, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.24445893089960888, | |
| "grad_norm": 9.288165092468262, | |
| "learning_rate": 0.0, | |
| "loss": 0.6639, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.24445893089960888, | |
| "eval_accuracy": 0.8080321285140563, | |
| "eval_f1": 0.8086136034722085, | |
| "eval_loss": 0.5128149390220642, | |
| "eval_runtime": 411.1581, | |
| "eval_samples_per_second": 6.056, | |
| "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 | |
| } | |