| { | |
| "best_metric": 0.4840739071369171, | |
| "best_model_checkpoint": "/mnt/beegfs/farid/mlora/outputs/xnli/aya-101/tr/rank4_lr5e-5/checkpoint-5500", | |
| "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": 1.8574336767196655, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.1457, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.020371577574967405, | |
| "eval_accuracy": 0.4429718875502008, | |
| "eval_f1": 0.4164069076365425, | |
| "eval_loss": 1.073067307472229, | |
| "eval_runtime": 413.6919, | |
| "eval_samples_per_second": 6.019, | |
| "eval_steps_per_second": 0.377, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.04074315514993481, | |
| "grad_norm": 10.016312599182129, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.9193, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.04074315514993481, | |
| "eval_accuracy": 0.738152610441767, | |
| "eval_f1": 0.7369047817418783, | |
| "eval_loss": 0.6893119812011719, | |
| "eval_runtime": 411.7464, | |
| "eval_samples_per_second": 6.047, | |
| "eval_steps_per_second": 0.379, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.06111473272490222, | |
| "grad_norm": 11.630990028381348, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.755, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.06111473272490222, | |
| "eval_accuracy": 0.770281124497992, | |
| "eval_f1": 0.7692403135749393, | |
| "eval_loss": 0.6047824025154114, | |
| "eval_runtime": 431.4567, | |
| "eval_samples_per_second": 5.771, | |
| "eval_steps_per_second": 0.362, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.08148631029986962, | |
| "grad_norm": 11.266816139221191, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.6935, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.08148631029986962, | |
| "eval_accuracy": 0.770281124497992, | |
| "eval_f1": 0.7716097568036839, | |
| "eval_loss": 0.5751612186431885, | |
| "eval_runtime": 411.4318, | |
| "eval_samples_per_second": 6.052, | |
| "eval_steps_per_second": 0.379, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.10185788787483703, | |
| "grad_norm": 5.681130886077881, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.6705, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.10185788787483703, | |
| "eval_accuracy": 0.7991967871485943, | |
| "eval_f1": 0.799397196834657, | |
| "eval_loss": 0.5203642845153809, | |
| "eval_runtime": 411.5367, | |
| "eval_samples_per_second": 6.05, | |
| "eval_steps_per_second": 0.379, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.12222946544980444, | |
| "grad_norm": 13.885538101196289, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.6535, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.12222946544980444, | |
| "eval_accuracy": 0.8080321285140563, | |
| "eval_f1": 0.8079778470492675, | |
| "eval_loss": 0.5077390670776367, | |
| "eval_runtime": 431.4507, | |
| "eval_samples_per_second": 5.771, | |
| "eval_steps_per_second": 0.362, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.14260104302477183, | |
| "grad_norm": 11.62227725982666, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.6344, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.14260104302477183, | |
| "eval_accuracy": 0.8056224899598393, | |
| "eval_f1": 0.8060120856906442, | |
| "eval_loss": 0.5052712559700012, | |
| "eval_runtime": 411.9037, | |
| "eval_samples_per_second": 6.045, | |
| "eval_steps_per_second": 0.379, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.16297262059973924, | |
| "grad_norm": 7.741968631744385, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.6461, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.16297262059973924, | |
| "eval_accuracy": 0.8168674698795181, | |
| "eval_f1": 0.8171764533567928, | |
| "eval_loss": 0.48436030745506287, | |
| "eval_runtime": 431.8438, | |
| "eval_samples_per_second": 5.766, | |
| "eval_steps_per_second": 0.361, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.18334419817470665, | |
| "grad_norm": 5.995279788970947, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.6149, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.18334419817470665, | |
| "eval_accuracy": 0.8072289156626506, | |
| "eval_f1": 0.8077366231036467, | |
| "eval_loss": 0.49473461508750916, | |
| "eval_runtime": 411.5781, | |
| "eval_samples_per_second": 6.05, | |
| "eval_steps_per_second": 0.379, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.20371577574967406, | |
| "grad_norm": 8.767409324645996, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.6235, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.20371577574967406, | |
| "eval_accuracy": 0.8052208835341366, | |
| "eval_f1": 0.8055606645944181, | |
| "eval_loss": 0.4961791932582855, | |
| "eval_runtime": 411.3835, | |
| "eval_samples_per_second": 6.053, | |
| "eval_steps_per_second": 0.379, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.22408735332464147, | |
| "grad_norm": 7.734911918640137, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.6145, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.22408735332464147, | |
| "eval_accuracy": 0.8112449799196787, | |
| "eval_f1": 0.8113855121951574, | |
| "eval_loss": 0.4840739071369171, | |
| "eval_runtime": 431.5261, | |
| "eval_samples_per_second": 5.77, | |
| "eval_steps_per_second": 0.362, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.24445893089960888, | |
| "grad_norm": 5.888601779937744, | |
| "learning_rate": 0.0, | |
| "loss": 0.6107, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.24445893089960888, | |
| "eval_accuracy": 0.8076305220883534, | |
| "eval_f1": 0.8079306262177152, | |
| "eval_loss": 0.48599937558174133, | |
| "eval_runtime": 411.7687, | |
| "eval_samples_per_second": 6.047, | |
| "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 | |
| } | |