| { | |
| "best_metric": 0.30846577882766724, | |
| "best_model_checkpoint": "uk-mt5-small-gec/checkpoint-750", | |
| "epoch": 1.950354609929078, | |
| "global_step": 1650, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0009822695035460994, | |
| "loss": 1.1542, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_google_bleu": 0.41727698590897433, | |
| "eval_loss": 0.40988656878471375, | |
| "eval_runtime": 182.204, | |
| "eval_samples_per_second": 16.493, | |
| "eval_steps_per_second": 0.516, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0009645390070921985, | |
| "loss": 0.4079, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_google_bleu": 0.417700589351588, | |
| "eval_loss": 0.3693748116493225, | |
| "eval_runtime": 184.6312, | |
| "eval_samples_per_second": 16.276, | |
| "eval_steps_per_second": 0.509, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.0009468085106382979, | |
| "loss": 0.3947, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_google_bleu": 0.42065639483903294, | |
| "eval_loss": 0.3382872939109802, | |
| "eval_runtime": 182.6943, | |
| "eval_samples_per_second": 16.448, | |
| "eval_steps_per_second": 0.515, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.0009290780141843972, | |
| "loss": 0.3699, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_google_bleu": 0.4213219269019765, | |
| "eval_loss": 0.329349160194397, | |
| "eval_runtime": 182.8557, | |
| "eval_samples_per_second": 16.434, | |
| "eval_steps_per_second": 0.514, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "learning_rate": 0.0009113475177304966, | |
| "loss": 0.384, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "eval_google_bleu": 0.41960295446754137, | |
| "eval_loss": 0.30846577882766724, | |
| "eval_runtime": 182.9587, | |
| "eval_samples_per_second": 16.424, | |
| "eval_steps_per_second": 0.514, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "learning_rate": 0.0008936170212765957, | |
| "loss": 0.301, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "eval_google_bleu": 0.41823031122631055, | |
| "eval_loss": 0.30877065658569336, | |
| "eval_runtime": 182.6834, | |
| "eval_samples_per_second": 16.449, | |
| "eval_steps_per_second": 0.515, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "learning_rate": 0.000875886524822695, | |
| "loss": 0.2613, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "eval_google_bleu": 0.41964051019514087, | |
| "eval_loss": 0.31575024127960205, | |
| "eval_runtime": 182.4274, | |
| "eval_samples_per_second": 16.472, | |
| "eval_steps_per_second": 0.515, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.0008581560283687944, | |
| "loss": 0.2715, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "eval_google_bleu": 0.41992548295187704, | |
| "eval_loss": 0.3233252167701721, | |
| "eval_runtime": 182.3422, | |
| "eval_samples_per_second": 16.48, | |
| "eval_steps_per_second": 0.516, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 0.0008404255319148936, | |
| "loss": 0.2861, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_google_bleu": 0.41982505463129355, | |
| "eval_loss": 0.31152355670928955, | |
| "eval_runtime": 182.3799, | |
| "eval_samples_per_second": 16.477, | |
| "eval_steps_per_second": 0.515, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.0008226950354609929, | |
| "loss": 0.2903, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_google_bleu": 0.41824821175792654, | |
| "eval_loss": 0.33445030450820923, | |
| "eval_runtime": 182.1203, | |
| "eval_samples_per_second": 16.5, | |
| "eval_steps_per_second": 0.516, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 0.0008049645390070922, | |
| "loss": 0.2811, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "eval_google_bleu": 0.42025291085487376, | |
| "eval_loss": 0.3121837377548218, | |
| "eval_runtime": 182.2332, | |
| "eval_samples_per_second": 16.49, | |
| "eval_steps_per_second": 0.516, | |
| "step": 1650 | |
| } | |
| ], | |
| "max_steps": 8460, | |
| "num_train_epochs": 10, | |
| "total_flos": 6751858532244480.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |