| { |
| "best_global_step": 3000, |
| "best_metric": 0.30037644505500793, |
| "best_model_checkpoint": "./emotion_model/checkpoint-3000", |
| "epoch": 3.0, |
| "eval_steps": 500, |
| "global_step": 3000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.1, |
| "grad_norm": 11.065900802612305, |
| "learning_rate": 1.934e-06, |
| "loss": 0.6485455322265625, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.2, |
| "grad_norm": 8.89755630493164, |
| "learning_rate": 1.8673333333333333e-06, |
| "loss": 0.6048553848266601, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.3, |
| "grad_norm": 4.825532913208008, |
| "learning_rate": 1.8006666666666666e-06, |
| "loss": 0.5499929809570312, |
| "step": 300 |
| }, |
| { |
| "epoch": 0.4, |
| "grad_norm": 8.28042984008789, |
| "learning_rate": 1.7339999999999998e-06, |
| "loss": 0.5557812881469727, |
| "step": 400 |
| }, |
| { |
| "epoch": 0.5, |
| "grad_norm": 6.205381393432617, |
| "learning_rate": 1.6673333333333333e-06, |
| "loss": 0.5276118087768554, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.6, |
| "grad_norm": 6.9054365158081055, |
| "learning_rate": 1.6006666666666666e-06, |
| "loss": 0.4956766510009766, |
| "step": 600 |
| }, |
| { |
| "epoch": 0.7, |
| "grad_norm": 7.5505757331848145, |
| "learning_rate": 1.534e-06, |
| "loss": 0.49066043853759767, |
| "step": 700 |
| }, |
| { |
| "epoch": 0.8, |
| "grad_norm": 7.9599223136901855, |
| "learning_rate": 1.4673333333333333e-06, |
| "loss": 0.4680234146118164, |
| "step": 800 |
| }, |
| { |
| "epoch": 0.9, |
| "grad_norm": 10.278119087219238, |
| "learning_rate": 1.4006666666666666e-06, |
| "loss": 0.43432594299316407, |
| "step": 900 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 13.048624038696289, |
| "learning_rate": 1.334e-06, |
| "loss": 0.4445612335205078, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.8825, |
| "eval_f1": 0.8777076002837825, |
| "eval_loss": 0.4004465341567993, |
| "eval_runtime": 8.5652, |
| "eval_samples_per_second": 233.504, |
| "eval_steps_per_second": 14.594, |
| "step": 1000 |
| }, |
| { |
| "epoch": 1.1, |
| "grad_norm": 5.21833610534668, |
| "learning_rate": 1.2673333333333333e-06, |
| "loss": 0.3828823471069336, |
| "step": 1100 |
| }, |
| { |
| "epoch": 1.2, |
| "grad_norm": 8.160884857177734, |
| "learning_rate": 1.2006666666666668e-06, |
| "loss": 0.3784292602539063, |
| "step": 1200 |
| }, |
| { |
| "epoch": 1.3, |
| "grad_norm": 5.752784729003906, |
| "learning_rate": 1.1339999999999999e-06, |
| "loss": 0.35453197479248044, |
| "step": 1300 |
| }, |
| { |
| "epoch": 1.4, |
| "grad_norm": 6.823087215423584, |
| "learning_rate": 1.0673333333333331e-06, |
| "loss": 0.36483436584472656, |
| "step": 1400 |
| }, |
| { |
| "epoch": 1.5, |
| "grad_norm": 2.917992115020752, |
| "learning_rate": 1.0006666666666666e-06, |
| "loss": 0.3727602005004883, |
| "step": 1500 |
| }, |
| { |
| "epoch": 1.6, |
| "grad_norm": 13.367361068725586, |
| "learning_rate": 9.34e-07, |
| "loss": 0.31851619720458985, |
| "step": 1600 |
| }, |
| { |
| "epoch": 1.7, |
| "grad_norm": 10.735733032226562, |
| "learning_rate": 8.673333333333332e-07, |
| "loss": 0.35459815979003906, |
| "step": 1700 |
| }, |
| { |
| "epoch": 1.8, |
| "grad_norm": 13.50684642791748, |
| "learning_rate": 8.006666666666666e-07, |
| "loss": 0.360652961730957, |
| "step": 1800 |
| }, |
| { |
| "epoch": 1.9, |
| "grad_norm": 6.711971282958984, |
| "learning_rate": 7.34e-07, |
| "loss": 0.33136638641357424, |
| "step": 1900 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 15.437868118286133, |
| "learning_rate": 6.673333333333334e-07, |
| "loss": 0.35294628143310547, |
| "step": 2000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8995, |
| "eval_f1": 0.8977802778475925, |
| "eval_loss": 0.3227323591709137, |
| "eval_runtime": 8.516, |
| "eval_samples_per_second": 234.853, |
| "eval_steps_per_second": 14.678, |
| "step": 2000 |
| }, |
| { |
| "epoch": 2.1, |
| "grad_norm": 12.73193073272705, |
| "learning_rate": 6.006666666666666e-07, |
| "loss": 0.320272331237793, |
| "step": 2100 |
| }, |
| { |
| "epoch": 2.2, |
| "grad_norm": 7.803280353546143, |
| "learning_rate": 5.34e-07, |
| "loss": 0.3086546516418457, |
| "step": 2200 |
| }, |
| { |
| "epoch": 2.3, |
| "grad_norm": 6.752017021179199, |
| "learning_rate": 4.673333333333333e-07, |
| "loss": 0.29211643218994143, |
| "step": 2300 |
| }, |
| { |
| "epoch": 2.4, |
| "grad_norm": 9.962376594543457, |
| "learning_rate": 4.0066666666666663e-07, |
| "loss": 0.29888750076293946, |
| "step": 2400 |
| }, |
| { |
| "epoch": 2.5, |
| "grad_norm": 3.3765363693237305, |
| "learning_rate": 3.34e-07, |
| "loss": 0.30436283111572265, |
| "step": 2500 |
| }, |
| { |
| "epoch": 2.6, |
| "grad_norm": 11.733818054199219, |
| "learning_rate": 2.673333333333333e-07, |
| "loss": 0.27375816345214843, |
| "step": 2600 |
| }, |
| { |
| "epoch": 2.7, |
| "grad_norm": 21.357440948486328, |
| "learning_rate": 2.0066666666666666e-07, |
| "loss": 0.3160665512084961, |
| "step": 2700 |
| }, |
| { |
| "epoch": 2.8, |
| "grad_norm": 43.54855728149414, |
| "learning_rate": 1.34e-07, |
| "loss": 0.28452653884887696, |
| "step": 2800 |
| }, |
| { |
| "epoch": 2.9, |
| "grad_norm": 15.32112979888916, |
| "learning_rate": 6.733333333333333e-08, |
| "loss": 0.32690986633300784, |
| "step": 2900 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 13.583910942077637, |
| "learning_rate": 6.666666666666666e-10, |
| "loss": 0.28569999694824216, |
| "step": 3000 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.9025, |
| "eval_f1": 0.9011700732312877, |
| "eval_loss": 0.30037644505500793, |
| "eval_runtime": 8.4939, |
| "eval_samples_per_second": 235.464, |
| "eval_steps_per_second": 14.717, |
| "step": 3000 |
| } |
| ], |
| "logging_steps": 100, |
| "max_steps": 3000, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1589722177536000.0, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
