| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 180, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05555555555555555, | |
| "grad_norm": 7.355470178103522, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 0.9936, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.2777777777777778, | |
| "grad_norm": 9.150698239786399, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.9935, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 3.0603212678780776, | |
| "learning_rate": 9.986966157589751e-06, | |
| "loss": 0.918, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 2.5047426466184493, | |
| "learning_rate": 9.934132612707631e-06, | |
| "loss": 0.8719, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 0.5942633748054504, | |
| "eval_runtime": 48.0382, | |
| "eval_samples_per_second": 11.532, | |
| "eval_steps_per_second": 0.375, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 2.854633191400044, | |
| "learning_rate": 9.841114703012817e-06, | |
| "loss": 0.711, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.3888888888888888, | |
| "grad_norm": 3.0542285095627815, | |
| "learning_rate": 9.708669981232542e-06, | |
| "loss": 0.5253, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 2.9849351058447633, | |
| "learning_rate": 9.537877098354787e-06, | |
| "loss": 0.5337, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.9444444444444444, | |
| "grad_norm": 2.6887676315185307, | |
| "learning_rate": 9.330127018922195e-06, | |
| "loss": 0.4801, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.2952410876750946, | |
| "eval_runtime": 47.9311, | |
| "eval_samples_per_second": 11.558, | |
| "eval_steps_per_second": 0.376, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 5.591116487375825, | |
| "learning_rate": 9.08711169279446e-06, | |
| "loss": 0.2937, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 2.583058931916133, | |
| "learning_rate": 8.810810275638183e-06, | |
| "loss": 0.2401, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.7777777777777777, | |
| "grad_norm": 3.0860918384845974, | |
| "learning_rate": 8.503473010366713e-06, | |
| "loss": 0.2452, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 0.12752698361873627, | |
| "eval_runtime": 47.9454, | |
| "eval_samples_per_second": 11.555, | |
| "eval_steps_per_second": 0.375, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 3.0555555555555554, | |
| "grad_norm": 10.045301056606435, | |
| "learning_rate": 8.167602900802121e-06, | |
| "loss": 0.2086, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 3.9568837117288234, | |
| "learning_rate": 7.805935326811913e-06, | |
| "loss": 0.1144, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.611111111111111, | |
| "grad_norm": 2.319433762920475, | |
| "learning_rate": 7.421415766938098e-06, | |
| "loss": 0.1038, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 3.888888888888889, | |
| "grad_norm": 2.1464957865789307, | |
| "learning_rate": 7.017175809949044e-06, | |
| "loss": 0.098, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.04484181106090546, | |
| "eval_runtime": 48.0587, | |
| "eval_samples_per_second": 11.528, | |
| "eval_steps_per_second": 0.375, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 4.166666666666667, | |
| "grad_norm": 1.7528155962578806, | |
| "learning_rate": 6.5965076506799e-06, | |
| "loss": 0.0637, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 1.8998235047238219, | |
| "learning_rate": 6.162837277871553e-06, | |
| "loss": 0.0492, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.722222222222222, | |
| "grad_norm": 1.727143217884404, | |
| "learning_rate": 5.719696572370596e-06, | |
| "loss": 0.0494, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.419982551702697, | |
| "learning_rate": 5.270694542927089e-06, | |
| "loss": 0.0439, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 0.025857795029878616, | |
| "eval_runtime": 47.9969, | |
| "eval_samples_per_second": 11.542, | |
| "eval_steps_per_second": 0.375, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 5.277777777777778, | |
| "grad_norm": 1.5318777607270062, | |
| "learning_rate": 4.819487933850319e-06, | |
| "loss": 0.0278, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 5.555555555555555, | |
| "grad_norm": 1.5662579234139284, | |
| "learning_rate": 4.369751443898554e-06, | |
| "loss": 0.0273, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 5.833333333333333, | |
| "grad_norm": 1.2926901021879944, | |
| "learning_rate": 3.92514779894488e-06, | |
| "loss": 0.0306, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.014916397631168365, | |
| "eval_runtime": 48.0425, | |
| "eval_samples_per_second": 11.531, | |
| "eval_steps_per_second": 0.375, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 6.111111111111111, | |
| "grad_norm": 0.859749869255068, | |
| "learning_rate": 3.489297922152136e-06, | |
| "loss": 0.0222, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 6.388888888888889, | |
| "grad_norm": 1.0464247096475507, | |
| "learning_rate": 3.0657514445958055e-06, | |
| "loss": 0.0136, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 1.355335206119406, | |
| "learning_rate": 2.65795779650105e-06, | |
| "loss": 0.0152, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.944444444444445, | |
| "grad_norm": 1.0250847159574037, | |
| "learning_rate": 2.269238114531515e-06, | |
| "loss": 0.017, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 0.006789184175431728, | |
| "eval_runtime": 47.9791, | |
| "eval_samples_per_second": 11.547, | |
| "eval_steps_per_second": 0.375, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 7.222222222222222, | |
| "grad_norm": 0.5699491687319692, | |
| "learning_rate": 1.9027581939213852e-06, | |
| "loss": 0.0078, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.663596820501962, | |
| "learning_rate": 1.561502705732883e-06, | |
| "loss": 0.0053, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 7.777777777777778, | |
| "grad_norm": 0.8863531225772922, | |
| "learning_rate": 1.2482508892179884e-06, | |
| "loss": 0.006, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.002534537808969617, | |
| "eval_runtime": 48.0729, | |
| "eval_samples_per_second": 11.524, | |
| "eval_steps_per_second": 0.374, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 8.055555555555555, | |
| "grad_norm": 0.4661094231461166, | |
| "learning_rate": 9.65553917249648e-07, | |
| "loss": 0.0065, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 8.333333333333334, | |
| "grad_norm": 0.39663216492684406, | |
| "learning_rate": 7.157141191620548e-07, | |
| "loss": 0.0021, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 8.61111111111111, | |
| "grad_norm": 0.3373551351268207, | |
| "learning_rate": 5.007662302124671e-07, | |
| "loss": 0.0019, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "grad_norm": 0.3425926240340406, | |
| "learning_rate": 3.224608203719953e-07, | |
| "loss": 0.0025, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 0.0013950355350971222, | |
| "eval_runtime": 47.9782, | |
| "eval_samples_per_second": 11.547, | |
| "eval_steps_per_second": 0.375, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 9.166666666666666, | |
| "grad_norm": 0.2182536824166498, | |
| "learning_rate": 1.8225003740388546e-07, | |
| "loss": 0.0021, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 9.444444444444445, | |
| "grad_norm": 0.1196015045593264, | |
| "learning_rate": 8.127578033998663e-08, | |
| "loss": 0.0012, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 9.722222222222221, | |
| "grad_norm": 0.24750584137098056, | |
| "learning_rate": 2.0360399672478826e-08, | |
| "loss": 0.0013, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.1571976207810524, | |
| "learning_rate": 0.0, | |
| "loss": 0.0012, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.001282889163121581, | |
| "eval_runtime": 47.8096, | |
| "eval_samples_per_second": 11.588, | |
| "eval_steps_per_second": 0.376, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 180, | |
| "total_flos": 37688338022400.0, | |
| "train_loss": 0.18709226546602117, | |
| "train_runtime": 2578.874, | |
| "train_samples_per_second": 2.148, | |
| "train_steps_per_second": 0.07 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 180, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 37688338022400.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |