| { | |
| "best_metric": 0.8020384907722473, | |
| "best_model_checkpoint": "data/Mistral-7B_task-2_60-samples_config-1_full/checkpoint-40", | |
| "epoch": 13.91304347826087, | |
| "eval_steps": 500, | |
| "global_step": 80, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 1.3230090141296387, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.1905, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 1.2287449836730957, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.2196, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 1.0540266036987305, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.1617, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 1.1279163360595703, | |
| "eval_runtime": 16.1097, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 0.9146004915237427, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.1428, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 0.7109998464584351, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.0994, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 0.4272143542766571, | |
| "learning_rate": 4e-05, | |
| "loss": 1.0666, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "eval_loss": 1.003680944442749, | |
| "eval_runtime": 16.1109, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 0.38766780495643616, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.0104, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 0.30721986293792725, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.9435, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 0.35625705122947693, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.9806, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.9565217391304346, | |
| "eval_loss": 0.928659200668335, | |
| "eval_runtime": 16.1088, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.130434782608696, | |
| "grad_norm": 0.40311527252197266, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.8916, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 0.5318938493728638, | |
| "learning_rate": 8e-05, | |
| "loss": 0.8667, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "grad_norm": 0.591539204120636, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.8359, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.8393594622612, | |
| "eval_runtime": 16.1071, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.30591532588005066, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.8328, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.521739130434782, | |
| "grad_norm": 1.6555272340774536, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.7954, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.311841756105423, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.7617, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.8171389102935791, | |
| "eval_runtime": 16.1098, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "grad_norm": 0.23419636487960815, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.7236, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.22347405552864075, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.7613, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "grad_norm": 0.21089664101600647, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.7312, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 0.8051286339759827, | |
| "eval_runtime": 16.1079, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.22305962443351746, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.7033, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.608695652173913, | |
| "grad_norm": 0.27895626425743103, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.7106, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 0.2448945939540863, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.6691, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 0.8020384907722473, | |
| "eval_runtime": 16.1102, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.304347826086957, | |
| "grad_norm": 0.26784563064575195, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.693, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.26836273074150085, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.6465, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.3266744613647461, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.64, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.8044763207435608, | |
| "eval_runtime": 16.106, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.33106833696365356, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.6313, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 0.4843381643295288, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.5832, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.869565217391305, | |
| "eval_loss": 0.8195890784263611, | |
| "eval_runtime": 16.1055, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.41880542039871216, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.5853, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.391304347826088, | |
| "grad_norm": 0.4882275462150574, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.5221, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.4971458315849304, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.5397, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.91304347826087, | |
| "eval_loss": 0.8470047116279602, | |
| "eval_runtime": 16.1097, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 10.08695652173913, | |
| "grad_norm": 0.48096963763237, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.5337, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.9099116325378418, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.4532, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "grad_norm": 0.7013418078422546, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.439, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.956521739130435, | |
| "eval_loss": 0.8771161437034607, | |
| "eval_runtime": 16.11, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.6685683131217957, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.4144, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 11.478260869565217, | |
| "grad_norm": 1.9589290618896484, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.4425, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 0.8325996994972229, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.3596, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.8885238170623779, | |
| "eval_runtime": 16.1094, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 12.173913043478262, | |
| "grad_norm": 0.9106132388114929, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.3526, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 0.8948937654495239, | |
| "learning_rate": 8.961448216775954e-05, | |
| "loss": 0.305, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "grad_norm": 1.1627625226974487, | |
| "learning_rate": 8.874722443520899e-05, | |
| "loss": 0.3268, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 0.9615630507469177, | |
| "eval_runtime": 16.1369, | |
| "eval_samples_per_second": 0.744, | |
| "eval_steps_per_second": 0.744, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 13.217391304347826, | |
| "grad_norm": 0.9224812388420105, | |
| "learning_rate": 8.784975278258783e-05, | |
| "loss": 0.303, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 13.565217391304348, | |
| "grad_norm": 0.980361819267273, | |
| "learning_rate": 8.692276703129421e-05, | |
| "loss": 0.2701, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 1.3148335218429565, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.2584, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_loss": 1.0826709270477295, | |
| "eval_runtime": 16.1313, | |
| "eval_samples_per_second": 0.744, | |
| "eval_steps_per_second": 0.744, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "step": 80, | |
| "total_flos": 7.711052129107968e+16, | |
| "train_loss": 0.6812634752131999, | |
| "train_runtime": 2814.9752, | |
| "train_samples_per_second": 0.817, | |
| "train_steps_per_second": 0.089 | |
| } | |
| ], | |
| "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": 7.711052129107968e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |