| { | |
| "best_metric": 1.1532875299453735, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-3_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": 0.5216583609580994, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.7136, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 0.5978745222091675, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.679, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.4568397104740143, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.6551, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 1.6558550596237183, | |
| "eval_runtime": 11.2908, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 0.5046256184577942, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.6533, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 0.4672154188156128, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.643, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 0.3166911005973816, | |
| "learning_rate": 4e-05, | |
| "loss": 1.627, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "eval_loss": 1.5674506425857544, | |
| "eval_runtime": 11.2915, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 0.31834301352500916, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.5925, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 0.35160449147224426, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.5075, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 0.256206750869751, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.5266, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.9565217391304346, | |
| "eval_loss": 1.4604921340942383, | |
| "eval_runtime": 11.2903, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.130434782608696, | |
| "grad_norm": 0.25195157527923584, | |
| "learning_rate": 7.2e-05, | |
| "loss": 1.409, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 0.2694939970970154, | |
| "learning_rate": 8e-05, | |
| "loss": 1.4093, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "grad_norm": 0.2599346339702606, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 1.3591, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 1.3447412252426147, | |
| "eval_runtime": 11.2897, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.3230060636997223, | |
| "learning_rate": 9.6e-05, | |
| "loss": 1.356, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.521739130434782, | |
| "grad_norm": 0.3146461546421051, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 1.2682, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.4067225754261017, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 1.2063, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 1.2404855489730835, | |
| "eval_runtime": 11.2901, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "grad_norm": 0.2216719090938568, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 1.1337, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.23555250465869904, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 1.1807, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "grad_norm": 0.170230895280838, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 1.1295, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 1.1922208070755005, | |
| "eval_runtime": 11.292, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.18546925485134125, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 1.1102, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.608695652173913, | |
| "grad_norm": 0.2131475806236267, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 1.1112, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 0.19139915704727173, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 1.0536, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 1.1716761589050293, | |
| "eval_runtime": 11.2874, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.304347826086957, | |
| "grad_norm": 0.19308795034885406, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 1.1212, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.19684337079524994, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 1.0459, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.1981879472732544, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 1.0393, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.1584736108779907, | |
| "eval_runtime": 11.2916, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.31062644720077515, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 1.053, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 0.21347109973430634, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 1.032, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.869565217391305, | |
| "eval_loss": 1.1532875299453735, | |
| "eval_runtime": 11.291, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.21405188739299774, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 1.0103, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.391304347826088, | |
| "grad_norm": 0.24655935168266296, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.9713, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.24110516905784607, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 1.0131, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.91304347826087, | |
| "eval_loss": 1.1537190675735474, | |
| "eval_runtime": 11.2881, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 10.08695652173913, | |
| "grad_norm": 0.2326139211654663, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 1.0051, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.257344514131546, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.9713, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "grad_norm": 0.2970510721206665, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.9202, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.956521739130435, | |
| "eval_loss": 1.1631392240524292, | |
| "eval_runtime": 11.2927, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.3109723627567291, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.9017, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 11.478260869565217, | |
| "grad_norm": 0.29920050501823425, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.9748, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 0.3235422372817993, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.8682, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.1720443964004517, | |
| "eval_runtime": 11.2881, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 12.173913043478262, | |
| "grad_norm": 0.3394072353839874, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.8832, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 0.3726913630962372, | |
| "learning_rate": 8.961448216775954e-05, | |
| "loss": 0.7961, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "grad_norm": 0.4422052800655365, | |
| "learning_rate": 8.874722443520899e-05, | |
| "loss": 0.8608, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 1.181230902671814, | |
| "eval_runtime": 11.2905, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 13.217391304347826, | |
| "grad_norm": 0.4691925048828125, | |
| "learning_rate": 8.784975278258783e-05, | |
| "loss": 0.802, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 13.565217391304348, | |
| "grad_norm": 0.5078902840614319, | |
| "learning_rate": 8.692276703129421e-05, | |
| "loss": 0.8039, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 0.7388030886650085, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.7371, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_loss": 1.2313176393508911, | |
| "eval_runtime": 11.2898, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 14.26086956521739, | |
| "grad_norm": 0.6160270571708679, | |
| "learning_rate": 8.498316702566828e-05, | |
| "loss": 0.6666, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 14.608695652173914, | |
| "grad_norm": 0.6270957589149475, | |
| "learning_rate": 8.397206521307584e-05, | |
| "loss": 0.6691, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 14.956521739130435, | |
| "grad_norm": 0.6252912878990173, | |
| "learning_rate": 8.293447300593402e-05, | |
| "loss": 0.6739, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 14.956521739130435, | |
| "eval_loss": 1.3222523927688599, | |
| "eval_runtime": 11.2899, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 15.304347826086957, | |
| "grad_norm": 0.6960534453392029, | |
| "learning_rate": 8.18711994874345e-05, | |
| "loss": 0.6207, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 15.652173913043478, | |
| "grad_norm": 1.0011051893234253, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.493, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 1.0199934244155884, | |
| "learning_rate": 7.967094433018508e-05, | |
| "loss": 0.6784, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 1.2982969284057617, | |
| "eval_runtime": 11.2889, | |
| "eval_samples_per_second": 1.063, | |
| "eval_steps_per_second": 1.063, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "step": 92, | |
| "total_flos": 5.89510592209879e+16, | |
| "train_loss": 1.0921854013982026, | |
| "train_runtime": 2175.8207, | |
| "train_samples_per_second": 1.057, | |
| "train_steps_per_second": 0.115 | |
| } | |
| ], | |
| "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": 5.89510592209879e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |