| { | |
| "best_metric": 0.7818325161933899, | |
| "best_model_checkpoint": "data/Mistral-7B_task-2_60-samples_config-1_full_auto/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": 0.9298184514045715, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.1312, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 1.0290251970291138, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.1626, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 0.8979393839836121, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.1106, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 1.0844870805740356, | |
| "eval_runtime": 16.5293, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 0.7789692878723145, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.0965, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 0.6168501973152161, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.0562, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 0.3783280551433563, | |
| "learning_rate": 4e-05, | |
| "loss": 1.0327, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "eval_loss": 0.9768473505973816, | |
| "eval_runtime": 16.5303, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 0.38281410932540894, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.9853, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 0.32894083857536316, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.9221, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 0.39342889189720154, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.9473, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.9565217391304346, | |
| "eval_loss": 0.903071403503418, | |
| "eval_runtime": 16.5314, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.130434782608696, | |
| "grad_norm": 0.4177634119987488, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.8711, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 0.499880313873291, | |
| "learning_rate": 8e-05, | |
| "loss": 0.8385, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "grad_norm": 0.3561837375164032, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.8032, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.8144460320472717, | |
| "eval_runtime": 16.527, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.3197736442089081, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.8076, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.521739130434782, | |
| "grad_norm": 0.28995320200920105, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.7696, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.28071844577789307, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.7285, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.791588306427002, | |
| "eval_runtime": 16.5278, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "grad_norm": 0.26790791749954224, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.6934, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.24303163588047028, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.7353, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "grad_norm": 0.2272714227437973, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.6954, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 0.7828149795532227, | |
| "eval_runtime": 16.5281, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.26335760951042175, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.6753, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.608695652173913, | |
| "grad_norm": 0.3373420238494873, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.6747, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 0.2719545066356659, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.6329, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 0.7818325161933899, | |
| "eval_runtime": 16.5268, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.304347826086957, | |
| "grad_norm": 0.2844427824020386, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.6545, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.34022724628448486, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.6095, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.33848240971565247, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.6007, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.7875502109527588, | |
| "eval_runtime": 16.5236, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.3588227331638336, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.5893, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 0.5320178270339966, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.5438, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.869565217391305, | |
| "eval_loss": 0.8092446327209473, | |
| "eval_runtime": 16.5254, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 0.39454472064971924, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.5465, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.391304347826088, | |
| "grad_norm": 0.4255107641220093, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.482, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.6079005599021912, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.4987, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.91304347826087, | |
| "eval_loss": 0.8370413184165955, | |
| "eval_runtime": 16.5251, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 10.08695652173913, | |
| "grad_norm": 0.507662296295166, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.4925, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.5482168793678284, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.4221, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "grad_norm": 0.7469679117202759, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.4104, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.956521739130435, | |
| "eval_loss": 0.8622700572013855, | |
| "eval_runtime": 16.5268, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.6494352221488953, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.3814, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 11.478260869565217, | |
| "grad_norm": 0.9218623042106628, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.4102, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 1.0047892332077026, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.3387, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.9042803645133972, | |
| "eval_runtime": 16.5261, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 12.173913043478262, | |
| "grad_norm": 0.7699682712554932, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.3229, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.521739130434783, | |
| "grad_norm": 0.8643565773963928, | |
| "learning_rate": 8.961448216775954e-05, | |
| "loss": 0.2837, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "grad_norm": 0.9567080736160278, | |
| "learning_rate": 8.874722443520899e-05, | |
| "loss": 0.3103, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 12.869565217391305, | |
| "eval_loss": 0.951578676700592, | |
| "eval_runtime": 16.5282, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 13.217391304347826, | |
| "grad_norm": 0.7713665962219238, | |
| "learning_rate": 8.784975278258783e-05, | |
| "loss": 0.2747, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 13.565217391304348, | |
| "grad_norm": 0.7671521306037903, | |
| "learning_rate": 8.692276703129421e-05, | |
| "loss": 0.2509, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "grad_norm": 1.304969072341919, | |
| "learning_rate": 8.596699001693255e-05, | |
| "loss": 0.2409, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "eval_loss": 1.0892488956451416, | |
| "eval_runtime": 16.5304, | |
| "eval_samples_per_second": 0.726, | |
| "eval_steps_per_second": 0.726, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 13.91304347826087, | |
| "step": 80, | |
| "total_flos": 7.987971064909005e+16, | |
| "train_loss": 0.6485219461843371, | |
| "train_runtime": 2899.6788, | |
| "train_samples_per_second": 0.793, | |
| "train_steps_per_second": 0.086 | |
| } | |
| ], | |
| "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": 7.987971064909005e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |