| { | |
| "best_metric": 0.48655232787132263, | |
| "best_model_checkpoint": "data/Mistral-7B_task-2_60-samples_config-2_auto/checkpoint-17", | |
| "epoch": 12.869565217391305, | |
| "eval_steps": 500, | |
| "global_step": 37, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 1.4357483386993408, | |
| "learning_rate": 1e-05, | |
| "loss": 0.7838, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 1.4391192197799683, | |
| "learning_rate": 2e-05, | |
| "loss": 0.7431, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "eval_loss": 0.7448654174804688, | |
| "eval_runtime": 16.5584, | |
| "eval_samples_per_second": 0.725, | |
| "eval_steps_per_second": 0.725, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 1.2061665058135986, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7384, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "eval_loss": 0.6326039433479309, | |
| "eval_runtime": 16.5454, | |
| "eval_samples_per_second": 0.725, | |
| "eval_steps_per_second": 0.725, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 0.7879842519760132, | |
| "learning_rate": 6e-05, | |
| "loss": 0.6677, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 0.6294183731079102, | |
| "learning_rate": 8e-05, | |
| "loss": 0.5622, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "eval_loss": 0.5592716932296753, | |
| "eval_runtime": 16.543, | |
| "eval_samples_per_second": 0.725, | |
| "eval_steps_per_second": 0.725, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 0.6157659888267517, | |
| "learning_rate": 0.0001, | |
| "loss": 0.5178, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "eval_loss": 0.5218959450721741, | |
| "eval_runtime": 16.5412, | |
| "eval_samples_per_second": 0.725, | |
| "eval_steps_per_second": 0.725, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.6588655114173889, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.495, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.5574151277542114, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.4354, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.4904453754425049, | |
| "eval_runtime": 16.5379, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.4835604131221771, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.3931, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 0.48655232787132263, | |
| "eval_runtime": 16.5381, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.5084338188171387, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.3449, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 0.46271926164627075, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.3085, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 0.4900866746902466, | |
| "eval_runtime": 16.5369, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.4965374171733856, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.2647, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.5032010674476624, | |
| "eval_runtime": 16.5365, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.47166964411735535, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.2271, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "eval_loss": 0.5300421714782715, | |
| "eval_runtime": 16.5391, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.5242656469345093, | |
| "learning_rate": 9.24024048078213e-05, | |
| "loss": 0.1913, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.48334604501724243, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.1316, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "eval_loss": 0.6065511107444763, | |
| "eval_runtime": 16.5444, | |
| "eval_samples_per_second": 0.725, | |
| "eval_steps_per_second": 0.725, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.430191308259964, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.11, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "eval_loss": 0.6681571006774902, | |
| "eval_runtime": 16.5364, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.5526455640792847, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.0774, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 0.5069611072540283, | |
| "learning_rate": 8.345653031794292e-05, | |
| "loss": 0.0479, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "eval_loss": 0.7444794178009033, | |
| "eval_runtime": 16.5372, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 0.47765862941741943, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.0271, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 0.8260830044746399, | |
| "eval_runtime": 16.5379, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "step": 37, | |
| "total_flos": 7.417401732130406e+16, | |
| "train_loss": 0.3416548876746281, | |
| "train_runtime": 2693.8062, | |
| "train_samples_per_second": 0.854, | |
| "train_steps_per_second": 0.037 | |
| } | |
| ], | |
| "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": 7.417401732130406e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |