| { | |
| "best_metric": 0.6915614008903503, | |
| "best_model_checkpoint": "distilbert_lda_5_v1_rte/checkpoint-70", | |
| "epoch": 12.0, | |
| "eval_steps": 500, | |
| "global_step": 120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.5855181813240051, | |
| "learning_rate": 0.00098, | |
| "loss": 0.9449, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.7021801471710205, | |
| "eval_runtime": 0.1254, | |
| "eval_samples_per_second": 2209.06, | |
| "eval_steps_per_second": 15.95, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 0.4455162584781647, | |
| "learning_rate": 0.00096, | |
| "loss": 0.7042, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.6999943852424622, | |
| "eval_runtime": 0.1299, | |
| "eval_samples_per_second": 2133.087, | |
| "eval_steps_per_second": 15.401, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 0.8011177182197571, | |
| "learning_rate": 0.00094, | |
| "loss": 0.7004, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.691927969455719, | |
| "eval_runtime": 0.1312, | |
| "eval_samples_per_second": 2111.997, | |
| "eval_steps_per_second": 15.249, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.39011821150779724, | |
| "learning_rate": 0.00092, | |
| "loss": 0.6962, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6929433941841125, | |
| "eval_runtime": 0.1287, | |
| "eval_samples_per_second": 2152.24, | |
| "eval_steps_per_second": 15.54, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.33056044578552246, | |
| "learning_rate": 0.0009000000000000001, | |
| "loss": 0.6965, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6995571851730347, | |
| "eval_runtime": 0.1316, | |
| "eval_samples_per_second": 2104.975, | |
| "eval_steps_per_second": 15.198, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 0.3222827613353729, | |
| "learning_rate": 0.00088, | |
| "loss": 0.6965, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6917728781700134, | |
| "eval_runtime": 0.129, | |
| "eval_samples_per_second": 2146.95, | |
| "eval_steps_per_second": 15.501, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 0.2771111726760864, | |
| "learning_rate": 0.00086, | |
| "loss": 0.6947, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6915614008903503, | |
| "eval_runtime": 0.1383, | |
| "eval_samples_per_second": 2002.531, | |
| "eval_steps_per_second": 14.459, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.3152952194213867, | |
| "learning_rate": 0.00084, | |
| "loss": 0.6949, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.6938458681106567, | |
| "eval_runtime": 0.1301, | |
| "eval_samples_per_second": 2129.334, | |
| "eval_steps_per_second": 15.374, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 0.508091390132904, | |
| "learning_rate": 0.00082, | |
| "loss": 0.6948, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.699120044708252, | |
| "eval_runtime": 0.1297, | |
| "eval_samples_per_second": 2135.915, | |
| "eval_steps_per_second": 15.422, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.13915003836154938, | |
| "learning_rate": 0.0008, | |
| "loss": 0.6957, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.5270758122743683, | |
| "eval_loss": 0.6919844150543213, | |
| "eval_runtime": 0.1283, | |
| "eval_samples_per_second": 2159.75, | |
| "eval_steps_per_second": 15.594, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "grad_norm": 0.2669129967689514, | |
| "learning_rate": 0.0007800000000000001, | |
| "loss": 0.6941, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.6955804228782654, | |
| "eval_runtime": 0.1281, | |
| "eval_samples_per_second": 2162.358, | |
| "eval_steps_per_second": 15.613, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.15157711505889893, | |
| "learning_rate": 0.00076, | |
| "loss": 0.6944, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4729241877256318, | |
| "eval_loss": 0.6934651136398315, | |
| "eval_runtime": 0.1305, | |
| "eval_samples_per_second": 2122.658, | |
| "eval_steps_per_second": 15.326, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 120, | |
| "total_flos": 1979062935920640.0, | |
| "train_loss": 0.7172688285509745, | |
| "train_runtime": 49.8636, | |
| "train_samples_per_second": 2496.812, | |
| "train_steps_per_second": 10.027 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 5, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 5 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1979062935920640.0, | |
| "train_batch_size": 256, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |