| { |
| "best_metric": 0.32960984110832214, |
| "best_model_checkpoint": "data/Mistral-7B_task-3_60-samples_config-1_auto/checkpoint-23", |
| "epoch": 10.956521739130435, |
| "eval_steps": 500, |
| "global_step": 63, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.17391304347826086, |
| "grad_norm": 27.30118751525879, |
| "learning_rate": 4.000000000000001e-06, |
| "loss": 4.8562, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.34782608695652173, |
| "grad_norm": 9.776118278503418, |
| "learning_rate": 8.000000000000001e-06, |
| "loss": 2.405, |
| "step": 2 |
| }, |
| { |
| "epoch": 0.6956521739130435, |
| "grad_norm": 11.994278907775879, |
| "learning_rate": 1.6000000000000003e-05, |
| "loss": 3.22, |
| "step": 4 |
| }, |
| { |
| "epoch": 0.8695652173913043, |
| "eval_loss": 2.4917044639587402, |
| "eval_runtime": 13.9534, |
| "eval_samples_per_second": 0.86, |
| "eval_steps_per_second": 0.86, |
| "step": 5 |
| }, |
| { |
| "epoch": 1.0434782608695652, |
| "grad_norm": 13.66543197631836, |
| "learning_rate": 2.4e-05, |
| "loss": 2.6567, |
| "step": 6 |
| }, |
| { |
| "epoch": 1.391304347826087, |
| "grad_norm": 8.681872367858887, |
| "learning_rate": 3.2000000000000005e-05, |
| "loss": 1.8405, |
| "step": 8 |
| }, |
| { |
| "epoch": 1.7391304347826086, |
| "grad_norm": 7.171779632568359, |
| "learning_rate": 4e-05, |
| "loss": 1.4002, |
| "step": 10 |
| }, |
| { |
| "epoch": 1.9130434782608696, |
| "eval_loss": 0.7615411877632141, |
| "eval_runtime": 13.9534, |
| "eval_samples_per_second": 0.86, |
| "eval_steps_per_second": 0.86, |
| "step": 11 |
| }, |
| { |
| "epoch": 2.0869565217391304, |
| "grad_norm": 2.9103076457977295, |
| "learning_rate": 4.8e-05, |
| "loss": 0.9604, |
| "step": 12 |
| }, |
| { |
| "epoch": 2.4347826086956523, |
| "grad_norm": 2.459990978240967, |
| "learning_rate": 5.6000000000000006e-05, |
| "loss": 0.4514, |
| "step": 14 |
| }, |
| { |
| "epoch": 2.782608695652174, |
| "grad_norm": 1.24560546875, |
| "learning_rate": 6.400000000000001e-05, |
| "loss": 0.4395, |
| "step": 16 |
| }, |
| { |
| "epoch": 2.9565217391304346, |
| "eval_loss": 0.4034530818462372, |
| "eval_runtime": 13.9521, |
| "eval_samples_per_second": 0.86, |
| "eval_steps_per_second": 0.86, |
| "step": 17 |
| }, |
| { |
| "epoch": 3.130434782608696, |
| "grad_norm": 1.1898442506790161, |
| "learning_rate": 7.2e-05, |
| "loss": 0.5894, |
| "step": 18 |
| }, |
| { |
| "epoch": 3.4782608695652173, |
| "grad_norm": 0.9692240953445435, |
| "learning_rate": 8e-05, |
| "loss": 0.347, |
| "step": 20 |
| }, |
| { |
| "epoch": 3.8260869565217392, |
| "grad_norm": 0.6413742303848267, |
| "learning_rate": 8.800000000000001e-05, |
| "loss": 0.261, |
| "step": 22 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.32960984110832214, |
| "eval_runtime": 13.9526, |
| "eval_samples_per_second": 0.86, |
| "eval_steps_per_second": 0.86, |
| "step": 23 |
| }, |
| { |
| "epoch": 4.173913043478261, |
| "grad_norm": 0.9576577544212341, |
| "learning_rate": 9.6e-05, |
| "loss": 0.2553, |
| "step": 24 |
| }, |
| { |
| "epoch": 4.521739130434782, |
| "grad_norm": 0.7608456611633301, |
| "learning_rate": 9.999512620046522e-05, |
| "loss": 0.2433, |
| "step": 26 |
| }, |
| { |
| "epoch": 4.869565217391305, |
| "grad_norm": 1.0061907768249512, |
| "learning_rate": 9.995614150494293e-05, |
| "loss": 0.1449, |
| "step": 28 |
| }, |
| { |
| "epoch": 4.869565217391305, |
| "eval_loss": 0.3313424289226532, |
| "eval_runtime": 13.9506, |
| "eval_samples_per_second": 0.86, |
| "eval_steps_per_second": 0.86, |
| "step": 28 |
| }, |
| { |
| "epoch": 5.217391304347826, |
| "grad_norm": 0.7992334961891174, |
| "learning_rate": 9.987820251299122e-05, |
| "loss": 0.1861, |
| "step": 30 |
| }, |
| { |
| "epoch": 5.565217391304348, |
| "grad_norm": 0.7203567028045654, |
| "learning_rate": 9.976136999909156e-05, |
| "loss": 0.1375, |
| "step": 32 |
| }, |
| { |
| "epoch": 5.913043478260869, |
| "grad_norm": 0.7088717222213745, |
| "learning_rate": 9.96057350657239e-05, |
| "loss": 0.1037, |
| "step": 34 |
| }, |
| { |
| "epoch": 5.913043478260869, |
| "eval_loss": 0.351863831281662, |
| "eval_runtime": 13.9518, |
| "eval_samples_per_second": 0.86, |
| "eval_steps_per_second": 0.86, |
| "step": 34 |
| }, |
| { |
| "epoch": 6.260869565217392, |
| "grad_norm": 0.6949975490570068, |
| "learning_rate": 9.941141907232765e-05, |
| "loss": 0.0834, |
| "step": 36 |
| }, |
| { |
| "epoch": 6.608695652173913, |
| "grad_norm": 0.9134985208511353, |
| "learning_rate": 9.917857354066931e-05, |
| "loss": 0.0926, |
| "step": 38 |
| }, |
| { |
| "epoch": 6.956521739130435, |
| "grad_norm": 0.6616380214691162, |
| "learning_rate": 9.890738003669029e-05, |
| "loss": 0.0466, |
| "step": 40 |
| }, |
| { |
| "epoch": 6.956521739130435, |
| "eval_loss": 0.40392160415649414, |
| "eval_runtime": 13.9547, |
| "eval_samples_per_second": 0.86, |
| "eval_steps_per_second": 0.86, |
| "step": 40 |
| }, |
| { |
| "epoch": 7.304347826086957, |
| "grad_norm": 0.5956407189369202, |
| "learning_rate": 9.859805002892732e-05, |
| "loss": 0.0318, |
| "step": 42 |
| }, |
| { |
| "epoch": 7.6521739130434785, |
| "grad_norm": 0.7603738307952881, |
| "learning_rate": 9.825082472361557e-05, |
| "loss": 0.0241, |
| "step": 44 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 0.6106905341148376, |
| "learning_rate": 9.786597487660337e-05, |
| "loss": 0.034, |
| "step": 46 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.4575117826461792, |
| "eval_runtime": 13.9532, |
| "eval_samples_per_second": 0.86, |
| "eval_steps_per_second": 0.86, |
| "step": 46 |
| }, |
| { |
| "epoch": 8.347826086956522, |
| "grad_norm": 0.25566983222961426, |
| "learning_rate": 9.744380058222483e-05, |
| "loss": 0.0112, |
| "step": 48 |
| }, |
| { |
| "epoch": 8.695652173913043, |
| "grad_norm": 0.5697676539421082, |
| "learning_rate": 9.698463103929542e-05, |
| "loss": 0.0127, |
| "step": 50 |
| }, |
| { |
| "epoch": 8.869565217391305, |
| "eval_loss": 0.5227072834968567, |
| "eval_runtime": 13.9519, |
| "eval_samples_per_second": 0.86, |
| "eval_steps_per_second": 0.86, |
| "step": 51 |
| }, |
| { |
| "epoch": 9.043478260869565, |
| "grad_norm": 0.38373303413391113, |
| "learning_rate": 9.648882429441257e-05, |
| "loss": 0.0139, |
| "step": 52 |
| }, |
| { |
| "epoch": 9.391304347826088, |
| "grad_norm": 0.5550734400749207, |
| "learning_rate": 9.595676696276172e-05, |
| "loss": 0.0081, |
| "step": 54 |
| }, |
| { |
| "epoch": 9.73913043478261, |
| "grad_norm": 0.4108295440673828, |
| "learning_rate": 9.538887392664544e-05, |
| "loss": 0.0115, |
| "step": 56 |
| }, |
| { |
| "epoch": 9.91304347826087, |
| "eval_loss": 0.4965360164642334, |
| "eval_runtime": 13.9507, |
| "eval_samples_per_second": 0.86, |
| "eval_steps_per_second": 0.86, |
| "step": 57 |
| }, |
| { |
| "epoch": 10.08695652173913, |
| "grad_norm": 0.15245622396469116, |
| "learning_rate": 9.478558801197065e-05, |
| "loss": 0.005, |
| "step": 58 |
| }, |
| { |
| "epoch": 10.434782608695652, |
| "grad_norm": 0.26796048879623413, |
| "learning_rate": 9.414737964294636e-05, |
| "loss": 0.006, |
| "step": 60 |
| }, |
| { |
| "epoch": 10.782608695652174, |
| "grad_norm": 0.20634447038173676, |
| "learning_rate": 9.347474647526095e-05, |
| "loss": 0.0054, |
| "step": 62 |
| }, |
| { |
| "epoch": 10.956521739130435, |
| "eval_loss": 0.5121752023696899, |
| "eval_runtime": 13.9572, |
| "eval_samples_per_second": 0.86, |
| "eval_steps_per_second": 0.86, |
| "step": 63 |
| }, |
| { |
| "epoch": 10.956521739130435, |
| "step": 63, |
| "total_flos": 5.034262521498829e+16, |
| "train_loss": 0.547967553700483, |
| "train_runtime": 1877.1688, |
| "train_samples_per_second": 1.225, |
| "train_steps_per_second": 0.133 |
| } |
| ], |
| "logging_steps": 2, |
| "max_steps": 250, |
| "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": 5.034262521498829e+16, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
