| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 20.0, | |
| "eval_steps": 500, | |
| "global_step": 10560, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.8086087107658386, | |
| "learning_rate": 4.75e-05, | |
| "loss": 0.4434, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9455927655768658, | |
| "eval_f1": 0.7605633802816901, | |
| "eval_loss": 0.16302905976772308, | |
| "eval_precision": 0.679945054945055, | |
| "eval_recall": 0.862870424171993, | |
| "eval_runtime": 4.5797, | |
| "eval_samples_per_second": 204.163, | |
| "eval_steps_per_second": 3.275, | |
| "step": 528 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.7114465832710266, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.1462, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9566729603497963, | |
| "eval_f1": 0.8073836276083467, | |
| "eval_loss": 0.12938551604747772, | |
| "eval_precision": 0.7481408031730292, | |
| "eval_recall": 0.8768158047646717, | |
| "eval_runtime": 4.6537, | |
| "eval_samples_per_second": 200.915, | |
| "eval_steps_per_second": 3.223, | |
| "step": 1056 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.316323161125183, | |
| "learning_rate": 4.25e-05, | |
| "loss": 0.1183, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9568717082381, | |
| "eval_f1": 0.8116639914392724, | |
| "eval_loss": 0.13784636557102203, | |
| "eval_precision": 0.7521070897372335, | |
| "eval_recall": 0.8814642649622313, | |
| "eval_runtime": 4.903, | |
| "eval_samples_per_second": 190.701, | |
| "eval_steps_per_second": 3.059, | |
| "step": 1584 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.4418916702270508, | |
| "learning_rate": 4e-05, | |
| "loss": 0.1012, | |
| "step": 2112 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9596541786743515, | |
| "eval_f1": 0.8231144872490505, | |
| "eval_loss": 0.1358918398618698, | |
| "eval_precision": 0.7720101781170483, | |
| "eval_recall": 0.8814642649622313, | |
| "eval_runtime": 4.7436, | |
| "eval_samples_per_second": 197.109, | |
| "eval_steps_per_second": 3.162, | |
| "step": 2112 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.0950379371643066, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.0884, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9622379012222995, | |
| "eval_f1": 0.8348926802421575, | |
| "eval_loss": 0.12661471962928772, | |
| "eval_precision": 0.7929952953476216, | |
| "eval_recall": 0.8814642649622313, | |
| "eval_runtime": 4.6673, | |
| "eval_samples_per_second": 200.331, | |
| "eval_steps_per_second": 3.214, | |
| "step": 2640 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 1.270456314086914, | |
| "learning_rate": 3.5e-05, | |
| "loss": 0.0793, | |
| "step": 3168 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.9610454138924774, | |
| "eval_f1": 0.8404432132963989, | |
| "eval_loss": 0.1408655196428299, | |
| "eval_precision": 0.803070407623081, | |
| "eval_recall": 0.8814642649622313, | |
| "eval_runtime": 4.6946, | |
| "eval_samples_per_second": 199.164, | |
| "eval_steps_per_second": 3.195, | |
| "step": 3168 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 1.6711246967315674, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.072, | |
| "step": 3696 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.9588591871211368, | |
| "eval_f1": 0.8222222222222223, | |
| "eval_loss": 0.1545909196138382, | |
| "eval_precision": 0.7704418486541391, | |
| "eval_recall": 0.8814642649622313, | |
| "eval_runtime": 4.6637, | |
| "eval_samples_per_second": 200.485, | |
| "eval_steps_per_second": 3.216, | |
| "step": 3696 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.5584707260131836, | |
| "learning_rate": 3e-05, | |
| "loss": 0.067, | |
| "step": 4224 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9608466660041737, | |
| "eval_f1": 0.8334258745141587, | |
| "eval_loss": 0.14326535165309906, | |
| "eval_precision": 0.797979797979798, | |
| "eval_recall": 0.8721673445671121, | |
| "eval_runtime": 4.6647, | |
| "eval_samples_per_second": 200.444, | |
| "eval_steps_per_second": 3.216, | |
| "step": 4224 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.5288811326026917, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.0607, | |
| "step": 4752 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.9598529265626553, | |
| "eval_f1": 0.8312328767123287, | |
| "eval_loss": 0.1468406319618225, | |
| "eval_precision": 0.7864178330741317, | |
| "eval_recall": 0.8814642649622313, | |
| "eval_runtime": 4.6583, | |
| "eval_samples_per_second": 200.719, | |
| "eval_steps_per_second": 3.22, | |
| "step": 4752 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 1.041274070739746, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0562, | |
| "step": 5280 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.9612441617807811, | |
| "eval_f1": 0.8267029972752045, | |
| "eval_loss": 0.14973483979701996, | |
| "eval_precision": 0.7783478707029245, | |
| "eval_recall": 0.8814642649622313, | |
| "eval_runtime": 4.6532, | |
| "eval_samples_per_second": 200.937, | |
| "eval_steps_per_second": 3.224, | |
| "step": 5280 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 1.7796343564987183, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.0506, | |
| "step": 5808 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.9594554307860479, | |
| "eval_f1": 0.8332413031474324, | |
| "eval_loss": 0.1600087583065033, | |
| "eval_precision": 0.793792740662809, | |
| "eval_recall": 0.8768158047646717, | |
| "eval_runtime": 4.6756, | |
| "eval_samples_per_second": 199.975, | |
| "eval_steps_per_second": 3.208, | |
| "step": 5808 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.6123548746109009, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0483, | |
| "step": 6336 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9608466660041737, | |
| "eval_f1": 0.8328721638074155, | |
| "eval_loss": 0.15964019298553467, | |
| "eval_precision": 0.7950343370311674, | |
| "eval_recall": 0.8744915746658919, | |
| "eval_runtime": 4.6719, | |
| "eval_samples_per_second": 200.134, | |
| "eval_steps_per_second": 3.211, | |
| "step": 6336 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "grad_norm": 0.9525193572044373, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.0443, | |
| "step": 6864 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.96064791811587, | |
| "eval_f1": 0.8237547892720306, | |
| "eval_loss": 0.15955425798892975, | |
| "eval_precision": 0.7785825142265907, | |
| "eval_recall": 0.8744915746658919, | |
| "eval_runtime": 4.6803, | |
| "eval_samples_per_second": 199.774, | |
| "eval_steps_per_second": 3.205, | |
| "step": 6864 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "grad_norm": 0.5356388688087463, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.0421, | |
| "step": 7392 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.9612441617807811, | |
| "eval_f1": 0.8350857775318208, | |
| "eval_loss": 0.16503094136714935, | |
| "eval_precision": 0.7971473851030111, | |
| "eval_recall": 0.8768158047646717, | |
| "eval_runtime": 4.671, | |
| "eval_samples_per_second": 200.172, | |
| "eval_steps_per_second": 3.211, | |
| "step": 7392 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "grad_norm": 0.6785407662391663, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.0395, | |
| "step": 7920 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.9602504223392626, | |
| "eval_f1": 0.8284449363586054, | |
| "eval_loss": 0.16934077441692352, | |
| "eval_precision": 0.7908082408874801, | |
| "eval_recall": 0.8698431144683324, | |
| "eval_runtime": 4.6607, | |
| "eval_samples_per_second": 200.612, | |
| "eval_steps_per_second": 3.218, | |
| "step": 7920 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.843337893486023, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0375, | |
| "step": 8448 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9594554307860479, | |
| "eval_f1": 0.8336088154269972, | |
| "eval_loss": 0.17250221967697144, | |
| "eval_precision": 0.7925615505500262, | |
| "eval_recall": 0.8791400348634515, | |
| "eval_runtime": 4.6872, | |
| "eval_samples_per_second": 199.478, | |
| "eval_steps_per_second": 3.2, | |
| "step": 8448 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "grad_norm": 0.6304071545600891, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.0358, | |
| "step": 8976 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.9612441617807811, | |
| "eval_f1": 0.8321289605336298, | |
| "eval_loss": 0.17892615497112274, | |
| "eval_precision": 0.7975492807671817, | |
| "eval_recall": 0.8698431144683324, | |
| "eval_runtime": 4.6497, | |
| "eval_samples_per_second": 201.086, | |
| "eval_steps_per_second": 3.226, | |
| "step": 8976 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "grad_norm": 1.1330559253692627, | |
| "learning_rate": 5e-06, | |
| "loss": 0.0339, | |
| "step": 9504 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.960051674450959, | |
| "eval_f1": 0.8225895316804408, | |
| "eval_loss": 0.17817425727844238, | |
| "eval_precision": 0.7820848611838659, | |
| "eval_recall": 0.8675188843695526, | |
| "eval_runtime": 4.6927, | |
| "eval_samples_per_second": 199.247, | |
| "eval_steps_per_second": 3.196, | |
| "step": 9504 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "grad_norm": 0.08901867270469666, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.0327, | |
| "step": 10032 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.9620391533339958, | |
| "eval_f1": 0.833983286908078, | |
| "eval_loss": 0.17433172464370728, | |
| "eval_precision": 0.8009630818619583, | |
| "eval_recall": 0.8698431144683324, | |
| "eval_runtime": 4.6614, | |
| "eval_samples_per_second": 200.583, | |
| "eval_steps_per_second": 3.218, | |
| "step": 10032 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 1.3878382444381714, | |
| "learning_rate": 0.0, | |
| "loss": 0.0327, | |
| "step": 10560 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9608466660041737, | |
| "eval_f1": 0.8321289605336298, | |
| "eval_loss": 0.17778323590755463, | |
| "eval_precision": 0.7975492807671817, | |
| "eval_recall": 0.8698431144683324, | |
| "eval_runtime": 4.9618, | |
| "eval_samples_per_second": 188.439, | |
| "eval_steps_per_second": 3.023, | |
| "step": 10560 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 10560, | |
| "total_flos": 5062583230111038.0, | |
| "train_loss": 0.08150525255636736, | |
| "train_runtime": 1343.1683, | |
| "train_samples_per_second": 125.628, | |
| "train_steps_per_second": 7.862 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10560, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 5062583230111038.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |