| { | |
| "best_metric": 2.381305694580078, | |
| "best_model_checkpoint": "./outputs/checkpoint-1400", | |
| "epoch": 1.0200364298724955, | |
| "eval_steps": 100, | |
| "global_step": 1400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 0.0002, | |
| "loss": 2.8179, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_loss": 2.723731517791748, | |
| "eval_runtime": 208.4619, | |
| "eval_samples_per_second": 30.097, | |
| "eval_steps_per_second": 3.766, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6952, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "eval_loss": 2.6728808879852295, | |
| "eval_runtime": 204.3784, | |
| "eval_samples_per_second": 30.698, | |
| "eval_steps_per_second": 3.841, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6508, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 2.6387226581573486, | |
| "eval_runtime": 205.5464, | |
| "eval_samples_per_second": 30.524, | |
| "eval_steps_per_second": 3.819, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6247, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "eval_loss": 2.6044600009918213, | |
| "eval_runtime": 204.5115, | |
| "eval_samples_per_second": 30.678, | |
| "eval_steps_per_second": 3.838, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5814, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_loss": 2.5784292221069336, | |
| "eval_runtime": 204.5265, | |
| "eval_samples_per_second": 30.676, | |
| "eval_steps_per_second": 3.838, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5538, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 2.5523250102996826, | |
| "eval_runtime": 205.1078, | |
| "eval_samples_per_second": 30.589, | |
| "eval_steps_per_second": 3.827, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5382, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 2.5274579524993896, | |
| "eval_runtime": 204.8352, | |
| "eval_samples_per_second": 30.63, | |
| "eval_steps_per_second": 3.832, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0002, | |
| "loss": 2.5141, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 2.506831645965576, | |
| "eval_runtime": 204.4886, | |
| "eval_samples_per_second": 30.681, | |
| "eval_steps_per_second": 3.839, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4765, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_loss": 2.484386920928955, | |
| "eval_runtime": 204.734, | |
| "eval_samples_per_second": 30.645, | |
| "eval_steps_per_second": 3.834, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4729, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_loss": 2.4648232460021973, | |
| "eval_runtime": 204.5514, | |
| "eval_samples_per_second": 30.672, | |
| "eval_steps_per_second": 3.838, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 0.0002, | |
| "loss": 2.468, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_loss": 2.442662477493286, | |
| "eval_runtime": 204.6558, | |
| "eval_samples_per_second": 30.656, | |
| "eval_steps_per_second": 3.836, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4198, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_loss": 2.4222779273986816, | |
| "eval_runtime": 204.5507, | |
| "eval_samples_per_second": 30.672, | |
| "eval_steps_per_second": 3.838, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "learning_rate": 0.0002, | |
| "loss": 2.4326, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 2.405611753463745, | |
| "eval_runtime": 204.6008, | |
| "eval_samples_per_second": 30.665, | |
| "eval_steps_per_second": 3.837, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "learning_rate": 0.0002, | |
| "loss": 2.3782, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 2.381305694580078, | |
| "eval_runtime": 204.4744, | |
| "eval_samples_per_second": 30.684, | |
| "eval_steps_per_second": 3.839, | |
| "step": 1400 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 4116, | |
| "num_train_epochs": 3, | |
| "save_steps": 100, | |
| "total_flos": 3.994021987006464e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |