| { | |
| "best_metric": 0.58187335729599, | |
| "best_model_checkpoint": "data/Llama-31-8B_task-2_120-samples_config-2_auto/checkpoint-38", | |
| "epoch": 14.0, | |
| "eval_steps": 500, | |
| "global_step": 77, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 0.4114352762699127, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.9261, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.36285364627838135, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.8467, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.39619794487953186, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.9245, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "eval_loss": 0.887127697467804, | |
| "eval_runtime": 25.0122, | |
| "eval_samples_per_second": 0.96, | |
| "eval_steps_per_second": 0.96, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 0.39810460805892944, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.887, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 0.39191126823425293, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.8678, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 0.3597930073738098, | |
| "learning_rate": 4e-05, | |
| "loss": 0.8834, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 0.7983560562133789, | |
| "eval_runtime": 25.0057, | |
| "eval_samples_per_second": 0.96, | |
| "eval_steps_per_second": 0.96, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 0.30901816487312317, | |
| "learning_rate": 4.8e-05, | |
| "loss": 0.752, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 0.278992235660553, | |
| "learning_rate": 5.6000000000000006e-05, | |
| "loss": 0.7998, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 0.3216788172721863, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 0.7224, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "eval_loss": 0.7229611277580261, | |
| "eval_runtime": 25.0099, | |
| "eval_samples_per_second": 0.96, | |
| "eval_steps_per_second": 0.96, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 0.34014081954956055, | |
| "learning_rate": 7.2e-05, | |
| "loss": 0.6977, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 0.2702198028564453, | |
| "learning_rate": 8e-05, | |
| "loss": 0.6652, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.2623775601387024, | |
| "learning_rate": 8.800000000000001e-05, | |
| "loss": 0.6496, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.6428138613700867, | |
| "eval_runtime": 25.008, | |
| "eval_samples_per_second": 0.96, | |
| "eval_steps_per_second": 0.96, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 0.27580147981643677, | |
| "learning_rate": 9.6e-05, | |
| "loss": 0.6274, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 0.3490183651447296, | |
| "learning_rate": 9.999512620046522e-05, | |
| "loss": 0.5918, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 4.909090909090909, | |
| "eval_loss": 0.6048721671104431, | |
| "eval_runtime": 25.0083, | |
| "eval_samples_per_second": 0.96, | |
| "eval_steps_per_second": 0.96, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 5.090909090909091, | |
| "grad_norm": 0.28270548582077026, | |
| "learning_rate": 9.995614150494293e-05, | |
| "loss": 0.564, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 0.294200599193573, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.5366, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 5.818181818181818, | |
| "grad_norm": 0.29604193568229675, | |
| "learning_rate": 9.976136999909156e-05, | |
| "loss": 0.5484, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.5820669531822205, | |
| "eval_runtime": 25.0148, | |
| "eval_samples_per_second": 0.959, | |
| "eval_steps_per_second": 0.959, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 6.181818181818182, | |
| "grad_norm": 0.25547143816947937, | |
| "learning_rate": 9.96057350657239e-05, | |
| "loss": 0.5068, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 6.545454545454545, | |
| "grad_norm": 0.2587781846523285, | |
| "learning_rate": 9.941141907232765e-05, | |
| "loss": 0.5096, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "grad_norm": 0.30631354451179504, | |
| "learning_rate": 9.917857354066931e-05, | |
| "loss": 0.472, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 6.909090909090909, | |
| "eval_loss": 0.58187335729599, | |
| "eval_runtime": 25.0075, | |
| "eval_samples_per_second": 0.96, | |
| "eval_steps_per_second": 0.96, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 0.2765539586544037, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.4694, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 7.636363636363637, | |
| "grad_norm": 0.307478129863739, | |
| "learning_rate": 9.859805002892732e-05, | |
| "loss": 0.4066, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 0.3184185326099396, | |
| "learning_rate": 9.825082472361557e-05, | |
| "loss": 0.4328, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.5885212421417236, | |
| "eval_runtime": 25.006, | |
| "eval_samples_per_second": 0.96, | |
| "eval_steps_per_second": 0.96, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.363636363636363, | |
| "grad_norm": 0.3068244159221649, | |
| "learning_rate": 9.786597487660337e-05, | |
| "loss": 0.3683, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 8.727272727272727, | |
| "grad_norm": 0.3444615602493286, | |
| "learning_rate": 9.744380058222483e-05, | |
| "loss": 0.3688, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 8.909090909090908, | |
| "eval_loss": 0.6169032454490662, | |
| "eval_runtime": 25.0151, | |
| "eval_samples_per_second": 0.959, | |
| "eval_steps_per_second": 0.959, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 0.44522011280059814, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.3687, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 9.454545454545455, | |
| "grad_norm": 0.40375861525535583, | |
| "learning_rate": 9.648882429441257e-05, | |
| "loss": 0.2941, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 9.818181818181818, | |
| "grad_norm": 0.480668842792511, | |
| "learning_rate": 9.595676696276172e-05, | |
| "loss": 0.2923, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 0.6638128161430359, | |
| "eval_runtime": 25.0073, | |
| "eval_samples_per_second": 0.96, | |
| "eval_steps_per_second": 0.96, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 10.181818181818182, | |
| "grad_norm": 0.44117558002471924, | |
| "learning_rate": 9.538887392664544e-05, | |
| "loss": 0.246, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 10.545454545454545, | |
| "grad_norm": 0.492409884929657, | |
| "learning_rate": 9.478558801197065e-05, | |
| "loss": 0.2068, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 0.6178811192512512, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.2063, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "eval_loss": 0.7678647041320801, | |
| "eval_runtime": 25.0103, | |
| "eval_samples_per_second": 0.96, | |
| "eval_steps_per_second": 0.96, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 11.272727272727273, | |
| "grad_norm": 0.5061853528022766, | |
| "learning_rate": 9.347474647526095e-05, | |
| "loss": 0.1539, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 11.636363636363637, | |
| "grad_norm": 0.44892534613609314, | |
| "learning_rate": 9.276821300802534e-05, | |
| "loss": 0.1254, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 0.643571674823761, | |
| "learning_rate": 9.202833017478422e-05, | |
| "loss": 0.1206, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 0.858501136302948, | |
| "eval_runtime": 25.0147, | |
| "eval_samples_per_second": 0.959, | |
| "eval_steps_per_second": 0.959, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.363636363636363, | |
| "grad_norm": 0.45422667264938354, | |
| "learning_rate": 9.125567491391476e-05, | |
| "loss": 0.072, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 12.727272727272727, | |
| "grad_norm": 0.5882347226142883, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.0643, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.909090909090908, | |
| "eval_loss": 1.0240541696548462, | |
| "eval_runtime": 25.0111, | |
| "eval_samples_per_second": 0.96, | |
| "eval_steps_per_second": 0.96, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 13.090909090909092, | |
| "grad_norm": 0.4740930199623108, | |
| "learning_rate": 8.961448216775954e-05, | |
| "loss": 0.0645, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 13.454545454545455, | |
| "grad_norm": 0.4891839325428009, | |
| "learning_rate": 8.874722443520899e-05, | |
| "loss": 0.0418, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 13.818181818181818, | |
| "grad_norm": 0.6079157590866089, | |
| "learning_rate": 8.784975278258783e-05, | |
| "loss": 0.0373, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_loss": 1.0912126302719116, | |
| "eval_runtime": 25.0178, | |
| "eval_samples_per_second": 0.959, | |
| "eval_steps_per_second": 0.959, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "step": 77, | |
| "total_flos": 1.219390286677934e+17, | |
| "train_loss": 0.4532604143023491, | |
| "train_runtime": 4348.0981, | |
| "train_samples_per_second": 1.012, | |
| "train_steps_per_second": 0.057 | |
| } | |
| ], | |
| "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": 1.219390286677934e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |