| { | |
| "best_metric": 0.19492121040821075, | |
| "best_model_checkpoint": "xblock-social-screenshots/checkpoint-482", | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 482, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 9.764724731445312, | |
| "learning_rate": 7.5342465753424655e-06, | |
| "loss": 0.8744, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 6.759668827056885, | |
| "learning_rate": 1.5068493150684931e-05, | |
| "loss": 0.7068, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 5.702600479125977, | |
| "learning_rate": 2.328767123287671e-05, | |
| "loss": 0.4411, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 4.835807800292969, | |
| "learning_rate": 3.1506849315068496e-05, | |
| "loss": 0.3594, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 1.6052229404449463, | |
| "learning_rate": 3.9726027397260274e-05, | |
| "loss": 0.2137, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 10.121936798095703, | |
| "learning_rate": 4.794520547945205e-05, | |
| "loss": 0.2919, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 3.0559985637664795, | |
| "learning_rate": 4.930769230769231e-05, | |
| "loss": 0.2982, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 8.374061584472656, | |
| "learning_rate": 4.8384615384615386e-05, | |
| "loss": 0.3612, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 1.3100831508636475, | |
| "learning_rate": 4.7461538461538465e-05, | |
| "loss": 0.247, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 12.590871810913086, | |
| "learning_rate": 4.653846153846154e-05, | |
| "loss": 0.3005, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 0.9043705463409424, | |
| "learning_rate": 4.5615384615384615e-05, | |
| "loss": 0.399, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 0.23747485876083374, | |
| "learning_rate": 4.4692307692307693e-05, | |
| "loss": 0.1496, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 10.08144760131836, | |
| "learning_rate": 4.376923076923077e-05, | |
| "loss": 0.1989, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 6.865645885467529, | |
| "learning_rate": 4.284615384615385e-05, | |
| "loss": 0.3254, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 4.330836296081543, | |
| "learning_rate": 4.192307692307693e-05, | |
| "loss": 0.2285, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 3.9234697818756104, | |
| "learning_rate": 4.1e-05, | |
| "loss": 0.1921, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 1.4691423177719116, | |
| "learning_rate": 4.007692307692308e-05, | |
| "loss": 0.348, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 0.398327499628067, | |
| "learning_rate": 3.915384615384616e-05, | |
| "loss": 0.1605, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 3.6116676330566406, | |
| "learning_rate": 3.823076923076923e-05, | |
| "loss": 0.2432, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.39185717701911926, | |
| "learning_rate": 3.730769230769231e-05, | |
| "loss": 0.0795, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9337474120082816, | |
| "eval_auc": 0.9920519240002744, | |
| "eval_f1": 0.937007874015748, | |
| "eval_loss": 0.20392270386219025, | |
| "eval_precision": 0.8847583643122676, | |
| "eval_recall": 0.99581589958159, | |
| "eval_runtime": 34.2431, | |
| "eval_samples_per_second": 14.105, | |
| "eval_steps_per_second": 0.905, | |
| "step": 241 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 0.1400236040353775, | |
| "learning_rate": 3.6384615384615386e-05, | |
| "loss": 0.2248, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 7.511680603027344, | |
| "learning_rate": 3.5461538461538464e-05, | |
| "loss": 0.1798, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 0.11658162623643875, | |
| "learning_rate": 3.453846153846154e-05, | |
| "loss": 0.1071, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 7.7041544914245605, | |
| "learning_rate": 3.3615384615384615e-05, | |
| "loss": 0.0488, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 0.3671063184738159, | |
| "learning_rate": 3.269230769230769e-05, | |
| "loss": 0.1775, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 7.276589870452881, | |
| "learning_rate": 3.176923076923077e-05, | |
| "loss": 0.1291, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 0.03822338953614235, | |
| "learning_rate": 3.084615384615385e-05, | |
| "loss": 0.1253, | |
| "step": 324 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 0.15799850225448608, | |
| "learning_rate": 2.9923076923076925e-05, | |
| "loss": 0.2956, | |
| "step": 336 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 3.221668004989624, | |
| "learning_rate": 2.9e-05, | |
| "loss": 0.1433, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "grad_norm": 1.780090093612671, | |
| "learning_rate": 2.807692307692308e-05, | |
| "loss": 0.1282, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 3.4230823516845703, | |
| "learning_rate": 2.7153846153846157e-05, | |
| "loss": 0.1416, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 0.7402670383453369, | |
| "learning_rate": 2.6230769230769232e-05, | |
| "loss": 0.1831, | |
| "step": 384 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 0.3014300763607025, | |
| "learning_rate": 2.530769230769231e-05, | |
| "loss": 0.2397, | |
| "step": 396 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 0.061521705240011215, | |
| "learning_rate": 2.4384615384615386e-05, | |
| "loss": 0.1162, | |
| "step": 408 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 0.593765139579773, | |
| "learning_rate": 2.3461538461538464e-05, | |
| "loss": 0.0587, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 2.235006093978882, | |
| "learning_rate": 2.253846153846154e-05, | |
| "loss": 0.2774, | |
| "step": 432 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 7.07012939453125, | |
| "learning_rate": 2.1615384615384614e-05, | |
| "loss": 0.137, | |
| "step": 444 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "grad_norm": 0.4600817561149597, | |
| "learning_rate": 2.0692307692307693e-05, | |
| "loss": 0.2305, | |
| "step": 456 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 5.892112731933594, | |
| "learning_rate": 1.976923076923077e-05, | |
| "loss": 0.2878, | |
| "step": 468 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 8.57860279083252, | |
| "learning_rate": 1.8846153846153846e-05, | |
| "loss": 0.188, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9399585921325052, | |
| "eval_auc": 0.9930893751286096, | |
| "eval_f1": 0.9423459244532804, | |
| "eval_loss": 0.19492121040821075, | |
| "eval_precision": 0.8977272727272727, | |
| "eval_recall": 0.9916317991631799, | |
| "eval_runtime": 34.2263, | |
| "eval_samples_per_second": 14.112, | |
| "eval_steps_per_second": 0.906, | |
| "step": 482 | |
| } | |
| ], | |
| "logging_steps": 12, | |
| "max_steps": 723, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 2.9880911195386675e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |