| { | |
| "best_metric": 0.7822471261024475, | |
| "best_model_checkpoint": "data/Mistral-7B_task-2_60-samples_config-2_full_auto/checkpoint-25", | |
| "epoch": 16.0, | |
| "eval_steps": 500, | |
| "global_step": 46, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 0.9489189386367798, | |
| "learning_rate": 1e-05, | |
| "loss": 1.1469, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.8985427618026733, | |
| "learning_rate": 2e-05, | |
| "loss": 1.1178, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "eval_loss": 1.112937331199646, | |
| "eval_runtime": 16.5274, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 0.8011978268623352, | |
| "learning_rate": 4e-05, | |
| "loss": 1.1097, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "eval_loss": 1.0291565656661987, | |
| "eval_runtime": 16.5345, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 0.5375749468803406, | |
| "learning_rate": 6e-05, | |
| "loss": 1.0616, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 0.39012864232063293, | |
| "learning_rate": 8e-05, | |
| "loss": 0.9835, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "eval_loss": 0.9618045687675476, | |
| "eval_runtime": 16.5291, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 0.3669772744178772, | |
| "learning_rate": 0.0001, | |
| "loss": 0.927, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "eval_loss": 0.9095487594604492, | |
| "eval_runtime": 16.5307, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.388439416885376, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.9117, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.4800668954849243, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.8549, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.8466771245002747, | |
| "eval_runtime": 16.527, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.29176655411720276, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.7961, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 0.8089969754219055, | |
| "eval_runtime": 16.5342, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.2345418632030487, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.7562, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 0.2233332395553589, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.736, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 0.7969680428504944, | |
| "eval_runtime": 16.5253, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.18992193043231964, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.7303, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.7857106328010559, | |
| "eval_runtime": 16.5342, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.1720886379480362, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.712, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "eval_loss": 0.7822471261024475, | |
| "eval_runtime": 16.5278, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.1806679219007492, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.6953, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.1769927591085434, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.6783, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "eval_loss": 0.782662570476532, | |
| "eval_runtime": 16.5267, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.20159979164600372, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.6731, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "eval_loss": 0.783447802066803, | |
| "eval_runtime": 16.5271, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.22783416509628296, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.6314, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 0.20514577627182007, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.6412, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "eval_loss": 0.7850430607795715, | |
| "eval_runtime": 16.5309, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 0.22831229865550995, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.6011, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 0.7914543151855469, | |
| "eval_runtime": 16.5296, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 13.217391304347826, | |
| "grad_norm": 0.2632400095462799, | |
| "learning_rate": 7.795964517353735e-05, | |
| "loss": 0.6055, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 0.24505576491355896, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.582, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_loss": 0.8032048344612122, | |
| "eval_runtime": 16.5261, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 14.608695652173914, | |
| "grad_norm": 0.27307823300361633, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.5422, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 14.956521739130435, | |
| "eval_loss": 0.8101192116737366, | |
| "eval_runtime": 16.5261, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 15.304347826086957, | |
| "grad_norm": 0.28701141476631165, | |
| "learning_rate": 6.873032967079561e-05, | |
| "loss": 0.539, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 0.41058349609375, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.515, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.8303666114807129, | |
| "eval_runtime": 16.5291, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "step": 46, | |
| "total_flos": 9.129109787508736e+16, | |
| "train_loss": 0.7571987328322037, | |
| "train_runtime": 3313.1474, | |
| "train_samples_per_second": 0.694, | |
| "train_steps_per_second": 0.03 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 7, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 9.129109787508736e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |