| { | |
| "best_metric": 1.0304415225982666, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-2_60-samples_config-1_full_auto/checkpoint-51", | |
| "epoch": 16.0, | |
| "eval_steps": 500, | |
| "global_step": 92, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.17391304347826086, | |
| "grad_norm": 0.48074769973754883, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.5183, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 0.505738377571106, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.5413, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.41624435782432556, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.4989, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 1.5038822889328003, | |
| "eval_runtime": 13.341, | |
| "eval_samples_per_second": 0.899, | |
| "eval_steps_per_second": 0.899, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 0.39714130759239197, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.486, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 0.4315492808818817, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.4793, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 0.32971909642219543, | |
| "learning_rate": 4e-05, | |
| "loss": 1.4648, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "eval_loss": 1.4134074449539185, | |
| "eval_runtime": 13.3423, | |
| "eval_samples_per_second": 0.899, | |
| "eval_steps_per_second": 0.899, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 0.31074926257133484, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.4197, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 0.29383906722068787, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.3327, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 0.24202194809913635, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.3625, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.9565217391304346, | |
| "eval_loss": 1.3140901327133179, | |
| "eval_runtime": 13.3414, | |
| "eval_samples_per_second": 0.899, | |
| "eval_steps_per_second": 0.899, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.130434782608696, | |
| "grad_norm": 0.23953472077846527, | |
| "learning_rate": 7.2e-05, | |
| "loss": 1.2689, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 0.28205573558807373, | |
| "learning_rate": 8e-05, | |
| "loss": 1.2562, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "grad_norm": 0.26320216059684753, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 1.2061, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.2031677961349487, | |
| "eval_runtime": 13.341, | |
| "eval_samples_per_second": 0.899, | |
| "eval_steps_per_second": 0.899, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.2358018457889557, | |
| "learning_rate": 9.6e-05, | |
| "loss": 1.2057, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.521739130434782, | |
| "grad_norm": 0.25525736808776855, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 1.1367, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.2895725667476654, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 1.078, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 1.11207914352417, | |
| "eval_runtime": 13.3388, | |
| "eval_samples_per_second": 0.9, | |
| "eval_steps_per_second": 0.9, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "grad_norm": 0.22301903367042542, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.0003, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.24768684804439545, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 1.0534, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "grad_norm": 0.20074968039989471, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.9963, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 1.064919352531433, | |
| "eval_runtime": 13.3389, | |
| "eval_samples_per_second": 0.9, | |
| "eval_steps_per_second": 0.9, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.15861091017723083, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.9747, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.608695652173913, | |
| "grad_norm": 0.1833048015832901, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.9961, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 0.16213782131671906, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.9285, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 1.0478910207748413, | |
| "eval_runtime": 13.3386, | |
| "eval_samples_per_second": 0.9, | |
| "eval_steps_per_second": 0.9, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.304347826086957, | |
| "grad_norm": 0.16214072704315186, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.9914, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.16307196021080017, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.926, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.1733638346195221, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.9295, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.03597891330719, | |
| "eval_runtime": 13.3382, | |
| "eval_samples_per_second": 0.9, | |
| "eval_steps_per_second": 0.9, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.15826472640037537, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.93, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 0.19020269811153412, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.9178, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.869565217391305, | |
| "eval_loss": 1.0304415225982666, | |
| "eval_runtime": 13.3423, | |
| "eval_samples_per_second": 0.899, | |
| "eval_steps_per_second": 0.899, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.17504218220710754, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.8981, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.391304347826088, | |
| "grad_norm": 0.17990776896476746, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.8507, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.19232700765132904, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.9031, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.91304347826087, | |
| "eval_loss": 1.0313078165054321, | |
| "eval_runtime": 13.3388, | |
| "eval_samples_per_second": 0.9, | |
| "eval_steps_per_second": 0.9, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 10.08695652173913, | |
| "grad_norm": 0.20676805078983307, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.895, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.20512782037258148, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.8544, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "grad_norm": 0.24138331413269043, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.8224, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.956521739130435, | |
| "eval_loss": 1.0367116928100586, | |
| "eval_runtime": 13.3416, | |
| "eval_samples_per_second": 0.899, | |
| "eval_steps_per_second": 0.899, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.2503072917461395, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.7976, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 11.478260869565217, | |
| "grad_norm": 0.24119675159454346, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.8636, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 0.26098811626434326, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.7715, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.0481574535369873, | |
| "eval_runtime": 13.339, | |
| "eval_samples_per_second": 0.9, | |
| "eval_steps_per_second": 0.9, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 12.173913043478262, | |
| "grad_norm": 0.2629372477531433, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.7858, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 0.32179731130599976, | |
| "learning_rate": 8.961448216775954e-05, | |
| "loss": 0.7089, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "grad_norm": 0.3544606566429138, | |
| "learning_rate": 8.874722443520899e-05, | |
| "loss": 0.7775, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 1.0639878511428833, | |
| "eval_runtime": 13.3461, | |
| "eval_samples_per_second": 0.899, | |
| "eval_steps_per_second": 0.899, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 13.217391304347826, | |
| "grad_norm": 0.35958898067474365, | |
| "learning_rate": 8.784975278258783e-05, | |
| "loss": 0.7168, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 13.565217391304348, | |
| "grad_norm": 0.39202362298965454, | |
| "learning_rate": 8.692276703129421e-05, | |
| "loss": 0.7192, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 0.49485886096954346, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.6631, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_loss": 1.1064558029174805, | |
| "eval_runtime": 13.3403, | |
| "eval_samples_per_second": 0.9, | |
| "eval_steps_per_second": 0.9, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 14.26086956521739, | |
| "grad_norm": 0.593677818775177, | |
| "learning_rate": 8.498316702566828e-05, | |
| "loss": 0.6085, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 14.608695652173914, | |
| "grad_norm": 0.5239640474319458, | |
| "learning_rate": 8.397206521307584e-05, | |
| "loss": 0.6042, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 14.956521739130435, | |
| "grad_norm": 0.4839284420013428, | |
| "learning_rate": 8.293447300593402e-05, | |
| "loss": 0.6127, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 14.956521739130435, | |
| "eval_loss": 1.138205647468567, | |
| "eval_runtime": 13.3379, | |
| "eval_samples_per_second": 0.9, | |
| "eval_steps_per_second": 0.9, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 15.304347826086957, | |
| "grad_norm": 0.7064086198806763, | |
| "learning_rate": 8.18711994874345e-05, | |
| "loss": 0.5708, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 15.652173913043478, | |
| "grad_norm": 0.638302743434906, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.457, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.1838070154190063, | |
| "learning_rate": 7.967094433018508e-05, | |
| "loss": 0.6162, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.1659694910049438, | |
| "eval_runtime": 13.3387, | |
| "eval_samples_per_second": 0.9, | |
| "eval_steps_per_second": 0.9, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "step": 92, | |
| "total_flos": 7.404745111109632e+16, | |
| "train_loss": 0.9753571735775989, | |
| "train_runtime": 2647.2901, | |
| "train_samples_per_second": 0.869, | |
| "train_steps_per_second": 0.094 | |
| } | |
| ], | |
| "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": 7.404745111109632e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |