| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 95214, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 3e-05, | |
| "loss": 2.2074, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_accuracy": 0.6701050400733948, | |
| "eval_loss": 1.3863972425460815, | |
| "eval_runtime": 1030.5845, | |
| "eval_samples_per_second": 109.496, | |
| "eval_steps_per_second": 9.125, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 2.877048978688162e-05, | |
| "loss": 1.2979, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_accuracy": 0.7089281678199768, | |
| "eval_loss": 1.1698851585388184, | |
| "eval_runtime": 1023.3048, | |
| "eval_samples_per_second": 110.275, | |
| "eval_steps_per_second": 9.19, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 2.7540979573763234e-05, | |
| "loss": 1.1725, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_accuracy": 0.723913311958313, | |
| "eval_loss": 1.0914552211761475, | |
| "eval_runtime": 1021.0004, | |
| "eval_samples_per_second": 110.524, | |
| "eval_steps_per_second": 9.211, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2.6311223360202048e-05, | |
| "loss": 1.1066, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_accuracy": 0.7364969849586487, | |
| "eval_loss": 1.0439897775650024, | |
| "eval_runtime": 1023.0505, | |
| "eval_samples_per_second": 110.302, | |
| "eval_steps_per_second": 9.192, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 2.3351475380761303e-05, | |
| "loss": 1.0399, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_accuracy": 0.7490540146827698, | |
| "eval_loss": 0.9834885001182556, | |
| "eval_runtime": 1112.3584, | |
| "eval_samples_per_second": 101.447, | |
| "eval_steps_per_second": 6.341, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 2.1689427361606846e-05, | |
| "loss": 1.0012, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_accuracy": 0.7581638693809509, | |
| "eval_loss": 0.9408503770828247, | |
| "eval_runtime": 1113.6514, | |
| "eval_samples_per_second": 101.329, | |
| "eval_steps_per_second": 6.333, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "learning_rate": 2.0027046799831514e-05, | |
| "loss": 0.8971, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "eval_accuracy": 0.7663254737854004, | |
| "eval_loss": 0.9190825819969177, | |
| "eval_runtime": 1114.1162, | |
| "eval_samples_per_second": 101.287, | |
| "eval_steps_per_second": 6.331, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 1.8364998780677056e-05, | |
| "loss": 0.8405, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_accuracy": 0.7693473100662231, | |
| "eval_loss": 0.9057066440582275, | |
| "eval_runtime": 1056.7123, | |
| "eval_samples_per_second": 106.789, | |
| "eval_steps_per_second": 6.674, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 1.6702950761522602e-05, | |
| "loss": 0.834, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_accuracy": 0.775940477848053, | |
| "eval_loss": 0.8832579851150513, | |
| "eval_runtime": 1059.1014, | |
| "eval_samples_per_second": 106.548, | |
| "eval_steps_per_second": 6.659, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 1.5040570199747268e-05, | |
| "loss": 0.8212, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_accuracy": 0.7792636156082153, | |
| "eval_loss": 0.866415798664093, | |
| "eval_runtime": 1054.2119, | |
| "eval_samples_per_second": 107.042, | |
| "eval_steps_per_second": 6.69, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 1.3378854723213692e-05, | |
| "loss": 0.8133, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_accuracy": 0.7830652594566345, | |
| "eval_loss": 0.8482676148414612, | |
| "eval_runtime": 1054.5931, | |
| "eval_samples_per_second": 107.003, | |
| "eval_steps_per_second": 6.688, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "learning_rate": 1.1716474161438358e-05, | |
| "loss": 0.7947, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "eval_accuracy": 0.7862643599510193, | |
| "eval_loss": 0.8344442844390869, | |
| "eval_runtime": 1058.4611, | |
| "eval_samples_per_second": 106.612, | |
| "eval_steps_per_second": 6.663, | |
| "step": 60000 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "learning_rate": 1.0054758684904782e-05, | |
| "loss": 0.7446, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "eval_accuracy": 0.7890114784240723, | |
| "eval_loss": 0.8323635458946228, | |
| "eval_runtime": 1054.4779, | |
| "eval_samples_per_second": 107.015, | |
| "eval_steps_per_second": 6.689, | |
| "step": 65000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "learning_rate": 8.392378123129448e-06, | |
| "loss": 0.6463, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 2.21, | |
| "eval_accuracy": 0.7913243770599365, | |
| "eval_loss": 0.8241144418716431, | |
| "eval_runtime": 1057.5768, | |
| "eval_samples_per_second": 106.701, | |
| "eval_steps_per_second": 6.669, | |
| "step": 70000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 6.7306626465958724e-06, | |
| "loss": 0.6461, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "eval_accuracy": 0.792999267578125, | |
| "eval_loss": 0.8187506794929504, | |
| "eval_runtime": 1055.2791, | |
| "eval_samples_per_second": 106.934, | |
| "eval_steps_per_second": 6.684, | |
| "step": 75000 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "learning_rate": 5.068282084820538e-06, | |
| "loss": 0.6403, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 2.52, | |
| "eval_accuracy": 0.7952146530151367, | |
| "eval_loss": 0.8096941709518433, | |
| "eval_runtime": 1058.933, | |
| "eval_samples_per_second": 106.565, | |
| "eval_steps_per_second": 6.66, | |
| "step": 80000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "learning_rate": 3.406566608286962e-06, | |
| "loss": 0.6294, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 2.68, | |
| "eval_accuracy": 0.7971464991569519, | |
| "eval_loss": 0.8044777512550354, | |
| "eval_runtime": 1078.5713, | |
| "eval_samples_per_second": 104.625, | |
| "eval_steps_per_second": 6.539, | |
| "step": 85000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 1.744186046511628e-06, | |
| "loss": 0.6198, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "eval_accuracy": 0.7985377907752991, | |
| "eval_loss": 0.7970817685127258, | |
| "eval_runtime": 1091.9096, | |
| "eval_samples_per_second": 103.346, | |
| "eval_steps_per_second": 6.459, | |
| "step": 90000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "learning_rate": 8.213802735717294e-08, | |
| "loss": 0.6169, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.7989366054534912, | |
| "eval_loss": 0.794947624206543, | |
| "eval_runtime": 1091.2301, | |
| "eval_samples_per_second": 103.411, | |
| "eval_steps_per_second": 6.463, | |
| "step": 95000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 95214, | |
| "total_flos": 2.918866335440248e+17, | |
| "train_loss": 0.06961118817632395, | |
| "train_runtime": 8353.1581, | |
| "train_samples_per_second": 364.749, | |
| "train_steps_per_second": 11.399 | |
| } | |
| ], | |
| "max_steps": 95214, | |
| "num_train_epochs": 3, | |
| "total_flos": 2.918866335440248e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |