| { | |
| "best_metric": 0.5375192165374756, | |
| "best_model_checkpoint": "data/Mistral-7B_task-2_60-samples_config-1/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": 4.75992488861084, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.986, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.34782608695652173, | |
| "grad_norm": 2.840470552444458, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.0599, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.6956521739130435, | |
| "grad_norm": 2.509486436843872, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.9616, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.8695652173913043, | |
| "eval_loss": 0.8767836689949036, | |
| "eval_runtime": 16.1173, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0434782608695652, | |
| "grad_norm": 2.077070474624634, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.8995, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.391304347826087, | |
| "grad_norm": 1.53449547290802, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.8161, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.7391304347826086, | |
| "grad_norm": 1.0011566877365112, | |
| "learning_rate": 4e-05, | |
| "loss": 0.7463, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.9130434782608696, | |
| "eval_loss": 0.6408722400665283, | |
| "eval_runtime": 16.1143, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.0869565217391304, | |
| "grad_norm": 0.7578339576721191, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.6299, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.4347826086956523, | |
| "grad_norm": 0.6122817397117615, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.5522, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.782608695652174, | |
| "grad_norm": 0.5560686588287354, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.6311, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.9565217391304346, | |
| "eval_loss": 0.5873411297798157, | |
| "eval_runtime": 16.1188, | |
| "eval_samples_per_second": 0.744, | |
| "eval_steps_per_second": 0.744, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 3.130434782608696, | |
| "grad_norm": 0.5497525334358215, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.5468, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.4782608695652173, | |
| "grad_norm": 0.6230431199073792, | |
| "learning_rate": 8e-05, | |
| "loss": 0.5046, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.8260869565217392, | |
| "grad_norm": 0.45565032958984375, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.5306, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.5466979146003723, | |
| "eval_runtime": 16.1126, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 4.173913043478261, | |
| "grad_norm": 0.42829808592796326, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.4993, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.521739130434782, | |
| "grad_norm": 0.4515722990036011, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.4521, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "grad_norm": 0.5315717458724976, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.4454, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 4.869565217391305, | |
| "eval_loss": 0.5375192165374756, | |
| "eval_runtime": 16.1187, | |
| "eval_samples_per_second": 0.744, | |
| "eval_steps_per_second": 0.744, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.217391304347826, | |
| "grad_norm": 0.46274569630622864, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.3712, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.565217391304348, | |
| "grad_norm": 0.5427079796791077, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.369, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "grad_norm": 0.5853497385978699, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.3475, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 5.913043478260869, | |
| "eval_loss": 0.5507214665412903, | |
| "eval_runtime": 16.1147, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.260869565217392, | |
| "grad_norm": 0.5865023136138916, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.2732, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.608695652173913, | |
| "grad_norm": 0.7380216121673584, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.2582, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "grad_norm": 1.2702546119689941, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.2291, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.956521739130435, | |
| "eval_loss": 0.6328728199005127, | |
| "eval_runtime": 16.1187, | |
| "eval_samples_per_second": 0.744, | |
| "eval_steps_per_second": 0.744, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.304347826086957, | |
| "grad_norm": 0.7579282522201538, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.183, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 7.6521739130434785, | |
| "grad_norm": 0.8423174023628235, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.1438, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.2522237300872803, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.1245, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.7402812838554382, | |
| "eval_runtime": 16.1135, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.347826086956522, | |
| "grad_norm": 0.8192511200904846, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.077, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.695652173913043, | |
| "grad_norm": 1.1571916341781616, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.0548, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.869565217391305, | |
| "eval_loss": 0.8629714846611023, | |
| "eval_runtime": 16.1114, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 9.043478260869565, | |
| "grad_norm": 1.533800721168518, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.0623, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.391304347826088, | |
| "grad_norm": 1.2077471017837524, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.0261, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 9.73913043478261, | |
| "grad_norm": 0.8268634080886841, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.0302, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.91304347826087, | |
| "eval_loss": 0.8625869750976562, | |
| "eval_runtime": 16.1131, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 10.08695652173913, | |
| "grad_norm": 0.9339629411697388, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.0289, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.434782608695652, | |
| "grad_norm": 0.7875646352767944, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.0126, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.782608695652174, | |
| "grad_norm": 0.8727169632911682, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.014, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.956521739130435, | |
| "eval_loss": 0.9664008021354675, | |
| "eval_runtime": 16.1149, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 11.130434782608695, | |
| "grad_norm": 0.6830615997314453, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.0187, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 11.478260869565217, | |
| "grad_norm": 0.628609836101532, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.0121, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 11.826086956521738, | |
| "grad_norm": 0.7100380659103394, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.0152, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.9425068497657776, | |
| "eval_runtime": 16.1161, | |
| "eval_samples_per_second": 0.745, | |
| "eval_steps_per_second": 0.745, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 69, | |
| "total_flos": 6.60947324633088e+16, | |
| "train_loss": 0.3448392077740552, | |
| "train_runtime": 2412.1178, | |
| "train_samples_per_second": 0.954, | |
| "train_steps_per_second": 0.104 | |
| } | |
| ], | |
| "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": 6.60947324633088e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |