| { | |
| "best_metric": 0.3063213527202606, | |
| "best_model_checkpoint": "./fine-tuned/checkpoint-1000", | |
| "epoch": 1.6625103906899419, | |
| "eval_steps": 100, | |
| "global_step": 1000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0831255195344971, | |
| "grad_norm": 81058.0859375, | |
| "learning_rate": 4.896006655574044e-05, | |
| "loss": 1.2803, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.1662510390689942, | |
| "grad_norm": 80198.9375, | |
| "learning_rate": 4.792013311148087e-05, | |
| "loss": 0.6108, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1662510390689942, | |
| "eval_loss": 0.4418589174747467, | |
| "eval_runtime": 38.3971, | |
| "eval_samples_per_second": 12.866, | |
| "eval_steps_per_second": 1.615, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.24937655860349128, | |
| "grad_norm": 70821.9921875, | |
| "learning_rate": 4.68801996672213e-05, | |
| "loss": 0.5174, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.3325020781379884, | |
| "grad_norm": 75381.640625, | |
| "learning_rate": 4.5840266222961734e-05, | |
| "loss": 0.4743, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.3325020781379884, | |
| "eval_loss": 0.3926239609718323, | |
| "eval_runtime": 38.3601, | |
| "eval_samples_per_second": 12.878, | |
| "eval_steps_per_second": 1.616, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.41562759767248547, | |
| "grad_norm": 87610.4453125, | |
| "learning_rate": 4.480033277870216e-05, | |
| "loss": 0.4777, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.49875311720698257, | |
| "grad_norm": 74392.109375, | |
| "learning_rate": 4.3760399334442597e-05, | |
| "loss": 0.4536, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.49875311720698257, | |
| "eval_loss": 0.36712247133255005, | |
| "eval_runtime": 38.3686, | |
| "eval_samples_per_second": 12.875, | |
| "eval_steps_per_second": 1.616, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.5818786367414797, | |
| "grad_norm": 152670.1875, | |
| "learning_rate": 4.272046589018303e-05, | |
| "loss": 0.4386, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.6650041562759768, | |
| "grad_norm": 59777.48046875, | |
| "learning_rate": 4.1680532445923466e-05, | |
| "loss": 0.449, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.6650041562759768, | |
| "eval_loss": 0.3487951159477234, | |
| "eval_runtime": 38.1805, | |
| "eval_samples_per_second": 12.939, | |
| "eval_steps_per_second": 1.624, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7481296758104738, | |
| "grad_norm": 60153.67578125, | |
| "learning_rate": 4.06405990016639e-05, | |
| "loss": 0.4116, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.8312551953449709, | |
| "grad_norm": 60111.30078125, | |
| "learning_rate": 3.960066555740433e-05, | |
| "loss": 0.4145, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.8312551953449709, | |
| "eval_loss": 0.3375319242477417, | |
| "eval_runtime": 38.2231, | |
| "eval_samples_per_second": 12.924, | |
| "eval_steps_per_second": 1.622, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.914380714879468, | |
| "grad_norm": 60626.75, | |
| "learning_rate": 3.856073211314476e-05, | |
| "loss": 0.405, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.9975062344139651, | |
| "grad_norm": 51910.04296875, | |
| "learning_rate": 3.752079866888519e-05, | |
| "loss": 0.4186, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.9975062344139651, | |
| "eval_loss": 0.3285529315471649, | |
| "eval_runtime": 38.3187, | |
| "eval_samples_per_second": 12.892, | |
| "eval_steps_per_second": 1.618, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.0806317539484622, | |
| "grad_norm": 48182.8046875, | |
| "learning_rate": 3.6480865224625625e-05, | |
| "loss": 0.3925, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.1637572734829593, | |
| "grad_norm": 51930.34765625, | |
| "learning_rate": 3.544093178036606e-05, | |
| "loss": 0.3705, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.1637572734829593, | |
| "eval_loss": 0.3227428197860718, | |
| "eval_runtime": 38.1185, | |
| "eval_samples_per_second": 12.96, | |
| "eval_steps_per_second": 1.627, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.2468827930174564, | |
| "grad_norm": 60283.91015625, | |
| "learning_rate": 3.4400998336106495e-05, | |
| "loss": 0.3823, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.3300083125519535, | |
| "grad_norm": 55843.62109375, | |
| "learning_rate": 3.336106489184692e-05, | |
| "loss": 0.3763, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.3300083125519535, | |
| "eval_loss": 0.3174193501472473, | |
| "eval_runtime": 38.4517, | |
| "eval_samples_per_second": 12.847, | |
| "eval_steps_per_second": 1.612, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.4131338320864506, | |
| "grad_norm": 40623.88671875, | |
| "learning_rate": 3.232113144758736e-05, | |
| "loss": 0.3509, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.4962593516209477, | |
| "grad_norm": 57212.0546875, | |
| "learning_rate": 3.128119800332779e-05, | |
| "loss": 0.3624, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.4962593516209477, | |
| "eval_loss": 0.31218209862709045, | |
| "eval_runtime": 38.1558, | |
| "eval_samples_per_second": 12.947, | |
| "eval_steps_per_second": 1.625, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.5793848711554448, | |
| "grad_norm": 49043.25390625, | |
| "learning_rate": 3.0241264559068223e-05, | |
| "loss": 0.3656, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.6625103906899419, | |
| "grad_norm": 50702.92578125, | |
| "learning_rate": 2.9201331114808654e-05, | |
| "loss": 0.3741, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.6625103906899419, | |
| "eval_loss": 0.3063213527202606, | |
| "eval_runtime": 38.2912, | |
| "eval_samples_per_second": 12.901, | |
| "eval_steps_per_second": 1.619, | |
| "step": 1000 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 2404, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9742717291069440.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |