| { | |
| "best_metric": 0.3692365884780884, | |
| "best_model_checkpoint": "data/Mistral-7B_task-3_60-samples_config-1/checkpoint-28", | |
| "epoch": 12.0, | |
| "eval_steps": 500, | |
| "global_step": 69, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 26.579113006591797, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 4.8883, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 10.55550765991211, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 2.7643, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 12.842199325561523, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 3.5059, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 2.7688920497894287, | |
| "eval_runtime": 13.809, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 13.838181495666504, | |
| "learning_rate": 2.4e-05, | |
| "loss": 2.9216, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 9.030895233154297, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 2.1254, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 7.164538860321045, | |
| "learning_rate": 4e-05, | |
| "loss": 1.5063, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "eval_loss": 0.8906886577606201, | |
| "eval_runtime": 13.8055, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 3.3004262447357178, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.0194, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 2.474790096282959, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.5294, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 1.3314909934997559, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.5013, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.9565217391304346, | |
| "eval_loss": 0.48438385128974915, | |
| "eval_runtime": 13.8088, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.130434782608696, | |
| "grad_norm": 0.9864939451217651, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.6322, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 1.1569544076919556, | |
| "learning_rate": 8e-05, | |
| "loss": 0.4462, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "grad_norm": 1.4715145826339722, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.2965, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.3944011628627777, | |
| "eval_runtime": 13.805, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.8321774005889893, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.2635, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.521739130434782, | |
| "grad_norm": 0.8808834552764893, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.3162, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 1.7142784595489502, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.189, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.3692365884780884, | |
| "eval_runtime": 13.8068, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "grad_norm": 1.060703158378601, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.2177, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.9269857406616211, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.1598, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "grad_norm": 0.7429260015487671, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.0884, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 0.46472927927970886, | |
| "eval_runtime": 13.807, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.5996131896972656, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.0693, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.608695652173913, | |
| "grad_norm": 1.4238375425338745, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.0818, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 1.3818721771240234, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.0487, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 0.6312050223350525, | |
| "eval_runtime": 13.8071, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.304347826086957, | |
| "grad_norm": 1.2610231637954712, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.0251, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.36835718154907227, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.0116, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.2624080181121826, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.0415, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.7006470561027527, | |
| "eval_runtime": 13.8109, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.5522323846817017, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.0056, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 0.5086768865585327, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0096, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.869565217391305, | |
| "eval_loss": 0.651463508605957, | |
| "eval_runtime": 13.8096, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.3935314118862152, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.0101, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.391304347826088, | |
| "grad_norm": 1.7814041376113892, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.003, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 2.2537779808044434, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.0226, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.91304347826087, | |
| "eval_loss": 0.8800778388977051, | |
| "eval_runtime": 13.8105, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 10.08695652173913, | |
| "grad_norm": 0.18110968172550201, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.0052, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 8.422475814819336, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0997, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "grad_norm": 9.830265045166016, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.019, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.956521739130435, | |
| "eval_loss": 0.6534069180488586, | |
| "eval_runtime": 13.8089, | |
| "eval_samples_per_second": 0.869, | |
| "eval_steps_per_second": 0.869, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.29976409673690796, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.0128, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 11.478260869565217, | |
| "grad_norm": 0.5346822738647461, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.006, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 0.2869880199432373, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.0075, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.6281649470329285, | |
| "eval_runtime": 13.8224, | |
| "eval_samples_per_second": 0.868, | |
| "eval_steps_per_second": 0.868, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 69, | |
| "total_flos": 5.440195292194406e+16, | |
| "train_loss": 0.5514464169669815, | |
| "train_runtime": 2028.9438, | |
| "train_samples_per_second": 1.134, | |
| "train_steps_per_second": 0.123 | |
| } | |
| ], | |
| "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": 5.440195292194406e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |