| { |
| "best_global_step": 100, |
| "best_metric": 0.010165676474571228, |
| "best_model_checkpoint": "/teamspace/studios/this_studio/DATN/output/medgemma_finetuned/checkpoint-100", |
| "epoch": 5.778588807785888, |
| "eval_steps": 25, |
| "global_step": 150, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.38929440389294406, |
| "grad_norm": 3.8712122440338135, |
| "learning_rate": 9.473684210526316e-05, |
| "loss": 0.4851, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.7785888077858881, |
| "grad_norm": 1.850216269493103, |
| "learning_rate": 0.0002, |
| "loss": 0.0256, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.9732360097323601, |
| "eval_loss": 0.015751386061310768, |
| "eval_runtime": 123.684, |
| "eval_samples_per_second": 0.412, |
| "eval_steps_per_second": 0.21, |
| "step": 25 |
| }, |
| { |
| "epoch": 1.1557177615571776, |
| "grad_norm": 0.2004503309726715, |
| "learning_rate": 0.00019814838910603481, |
| "loss": 0.0148, |
| "step": 30 |
| }, |
| { |
| "epoch": 1.5450121654501217, |
| "grad_norm": 0.07818326354026794, |
| "learning_rate": 0.0001926621256821922, |
| "loss": 0.0094, |
| "step": 40 |
| }, |
| { |
| "epoch": 1.9343065693430657, |
| "grad_norm": 0.0575609989464283, |
| "learning_rate": 0.00018374437823092724, |
| "loss": 0.009, |
| "step": 50 |
| }, |
| { |
| "epoch": 1.9343065693430657, |
| "eval_loss": 0.010756579227745533, |
| "eval_runtime": 101.2153, |
| "eval_samples_per_second": 0.504, |
| "eval_steps_per_second": 0.257, |
| "step": 50 |
| }, |
| { |
| "epoch": 2.3114355231143553, |
| "grad_norm": 0.03460681438446045, |
| "learning_rate": 0.0001717253907188477, |
| "loss": 0.0083, |
| "step": 60 |
| }, |
| { |
| "epoch": 2.7007299270072993, |
| "grad_norm": 0.03570393845438957, |
| "learning_rate": 0.0001570502529101896, |
| "loss": 0.008, |
| "step": 70 |
| }, |
| { |
| "epoch": 2.895377128953771, |
| "eval_loss": 0.011468109674751759, |
| "eval_runtime": 97.9517, |
| "eval_samples_per_second": 0.521, |
| "eval_steps_per_second": 0.265, |
| "step": 75 |
| }, |
| { |
| "epoch": 3.0778588807785887, |
| "grad_norm": 0.0510949082672596, |
| "learning_rate": 0.00014026241770569197, |
| "loss": 0.0078, |
| "step": 80 |
| }, |
| { |
| "epoch": 3.4671532846715327, |
| "grad_norm": 0.04285643249750137, |
| "learning_rate": 0.00012198357587636957, |
| "loss": 0.0073, |
| "step": 90 |
| }, |
| { |
| "epoch": 3.8564476885644767, |
| "grad_norm": 0.034959107637405396, |
| "learning_rate": 0.00010289063347542726, |
| "loss": 0.0075, |
| "step": 100 |
| }, |
| { |
| "epoch": 3.8564476885644767, |
| "eval_loss": 0.010165676474571228, |
| "eval_runtime": 98.2376, |
| "eval_samples_per_second": 0.519, |
| "eval_steps_per_second": 0.265, |
| "step": 100 |
| }, |
| { |
| "epoch": 4.233576642335766, |
| "grad_norm": 0.048688922077417374, |
| "learning_rate": 8.369064450581373e-05, |
| "loss": 0.0069, |
| "step": 110 |
| }, |
| { |
| "epoch": 4.622871046228711, |
| "grad_norm": 0.032025065273046494, |
| "learning_rate": 6.509462714233195e-05, |
| "loss": 0.0069, |
| "step": 120 |
| }, |
| { |
| "epoch": 4.817518248175182, |
| "eval_loss": 0.01066823210567236, |
| "eval_runtime": 98.7276, |
| "eval_samples_per_second": 0.517, |
| "eval_steps_per_second": 0.263, |
| "step": 125 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 0.04177581146359444, |
| "learning_rate": 4.779123315167362e-05, |
| "loss": 0.0073, |
| "step": 130 |
| }, |
| { |
| "epoch": 5.3892944038929445, |
| "grad_norm": 0.021832605823874474, |
| "learning_rate": 3.242124559015234e-05, |
| "loss": 0.0068, |
| "step": 140 |
| }, |
| { |
| "epoch": 5.778588807785888, |
| "grad_norm": 0.025266585871577263, |
| "learning_rate": 1.9553849185948512e-05, |
| "loss": 0.0068, |
| "step": 150 |
| }, |
| { |
| "epoch": 5.778588807785888, |
| "eval_loss": 0.010451124049723148, |
| "eval_runtime": 97.7624, |
| "eval_samples_per_second": 0.522, |
| "eval_steps_per_second": 0.266, |
| "step": 150 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 182, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 7, |
| "save_steps": 25, |
| "stateful_callbacks": { |
| "EarlyStoppingCallback": { |
| "args": { |
| "early_stopping_patience": 50, |
| "early_stopping_threshold": 0.0001 |
| }, |
| "attributes": { |
| "early_stopping_patience_counter": 2 |
| } |
| }, |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.2357930956632294e+17, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
