| { | |
| "best_metric": 0.8263590931892395, | |
| "best_model_checkpoint": "data/Mistral-7B_task-1_60-samples_config-1_full_auto/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": 2.075784206390381, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.7064, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 2.541001081466675, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.9196, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 2.1010255813598633, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.7884, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 1.6039167642593384, | |
| "eval_runtime": 5.1243, | |
| "eval_samples_per_second": 2.342, | |
| "eval_steps_per_second": 2.342, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 1.7213844060897827, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.7066, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 1.7199276685714722, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.5831, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 1.1836798191070557, | |
| "learning_rate": 4e-05, | |
| "loss": 1.4942, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "eval_loss": 1.293893575668335, | |
| "eval_runtime": 5.1275, | |
| "eval_samples_per_second": 2.34, | |
| "eval_steps_per_second": 2.34, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 1.3160754442214966, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.3983, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 1.6250238418579102, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 1.2286, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 1.2760955095291138, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 1.0868, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.9565217391304346, | |
| "eval_loss": 0.9507457613945007, | |
| "eval_runtime": 5.1265, | |
| "eval_samples_per_second": 2.341, | |
| "eval_steps_per_second": 2.341, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.130434782608696, | |
| "grad_norm": 1.0798264741897583, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.9799, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 1.2932850122451782, | |
| "learning_rate": 8e-05, | |
| "loss": 0.8436, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "grad_norm": 1.0376642942428589, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.809, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.864802896976471, | |
| "eval_runtime": 5.1274, | |
| "eval_samples_per_second": 2.34, | |
| "eval_steps_per_second": 2.34, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.9833399057388306, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.8034, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.521739130434782, | |
| "grad_norm": 0.7875859141349792, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.7396, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.8038161396980286, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.6926, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.8263590931892395, | |
| "eval_runtime": 5.1289, | |
| "eval_samples_per_second": 2.34, | |
| "eval_steps_per_second": 2.34, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "grad_norm": 0.7411243915557861, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.6336, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.895855724811554, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.5994, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "grad_norm": 1.0499613285064697, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.5667, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 0.8804253935813904, | |
| "eval_runtime": 5.1275, | |
| "eval_samples_per_second": 2.34, | |
| "eval_steps_per_second": 2.34, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.9823373556137085, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.469, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.608695652173913, | |
| "grad_norm": 1.1583163738250732, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.4567, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 1.0824776887893677, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.3857, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 0.9485516548156738, | |
| "eval_runtime": 5.1278, | |
| "eval_samples_per_second": 2.34, | |
| "eval_steps_per_second": 2.34, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.304347826086957, | |
| "grad_norm": 1.1419284343719482, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.3406, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 1.2314784526824951, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.2936, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.6639622449874878, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.2768, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 1.1056171655654907, | |
| "eval_runtime": 5.1218, | |
| "eval_samples_per_second": 2.343, | |
| "eval_steps_per_second": 2.343, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 1.286566138267517, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.2192, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 1.0422497987747192, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.2153, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.869565217391305, | |
| "eval_loss": 1.1607187986373901, | |
| "eval_runtime": 5.1285, | |
| "eval_samples_per_second": 2.34, | |
| "eval_steps_per_second": 2.34, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 1.0675088167190552, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.1865, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.391304347826088, | |
| "grad_norm": 1.297598123550415, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.1652, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 1.1824018955230713, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.1746, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.91304347826087, | |
| "eval_loss": 1.1830960512161255, | |
| "eval_runtime": 5.126, | |
| "eval_samples_per_second": 2.341, | |
| "eval_steps_per_second": 2.341, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 10.08695652173913, | |
| "grad_norm": 0.8748708367347717, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.148, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.9217429161071777, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.14, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "grad_norm": 1.0132285356521606, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.136, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.956521739130435, | |
| "eval_loss": 1.1961917877197266, | |
| "eval_runtime": 5.1268, | |
| "eval_samples_per_second": 2.341, | |
| "eval_steps_per_second": 2.341, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.7959060668945312, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.13, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 11.478260869565217, | |
| "grad_norm": 0.9177082777023315, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.1186, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 1.0011273622512817, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.1149, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.295631766319275, | |
| "eval_runtime": 5.1267, | |
| "eval_samples_per_second": 2.341, | |
| "eval_steps_per_second": 2.341, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 69, | |
| "total_flos": 8954933137113088.0, | |
| "train_loss": 0.6606330492574236, | |
| "train_runtime": 672.7478, | |
| "train_samples_per_second": 3.419, | |
| "train_steps_per_second": 0.372 | |
| } | |
| ], | |
| "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": 8954933137113088.0, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |