| { | |
| "best_metric": 0.7470817120622568, | |
| "best_model_checkpoint": "distilbert-base-multilingual-cased-hyper-matt/run-xafmgr6g/checkpoint-500", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 500, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 2.5906715393066406, | |
| "learning_rate": 4.508392080305315e-06, | |
| "loss": 0.6814, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 1.9834462404251099, | |
| "learning_rate": 4.41638407866643e-06, | |
| "loss": 0.6166, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 2.170398473739624, | |
| "learning_rate": 4.324376077027546e-06, | |
| "loss": 0.5797, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 2.0304064750671387, | |
| "learning_rate": 4.232368075388663e-06, | |
| "loss": 0.5878, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.8281610012054443, | |
| "learning_rate": 4.140360073749779e-06, | |
| "loss": 0.5843, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 3.344188928604126, | |
| "learning_rate": 4.048352072110894e-06, | |
| "loss": 0.5603, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 3.3330812454223633, | |
| "learning_rate": 3.9563440704720105e-06, | |
| "loss": 0.5264, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 2.7309162616729736, | |
| "learning_rate": 3.864336068833127e-06, | |
| "loss": 0.5039, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 2.202669620513916, | |
| "learning_rate": 3.7723280671942426e-06, | |
| "loss": 0.4878, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.3855249881744385, | |
| "learning_rate": 3.680320065555359e-06, | |
| "loss": 0.5067, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.705, | |
| "eval_f1": 0.41, | |
| "eval_loss": 0.5030852556228638, | |
| "eval_precision": 0.5324675324675324, | |
| "eval_recall": 0.3333333333333333, | |
| "eval_runtime": 1.5162, | |
| "eval_samples_per_second": 263.821, | |
| "eval_steps_per_second": 16.489, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 2.7073991298675537, | |
| "learning_rate": 3.588312063916475e-06, | |
| "loss": 0.4801, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 2.3832802772521973, | |
| "learning_rate": 3.496304062277591e-06, | |
| "loss": 0.5006, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 3.7265353202819824, | |
| "learning_rate": 3.4042960606387067e-06, | |
| "loss": 0.436, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 2.69901180267334, | |
| "learning_rate": 3.3122880589998226e-06, | |
| "loss": 0.4431, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 6.378077983856201, | |
| "learning_rate": 3.220280057360939e-06, | |
| "loss": 0.4665, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 6.366635799407959, | |
| "learning_rate": 3.128272055722055e-06, | |
| "loss": 0.4561, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 3.3578593730926514, | |
| "learning_rate": 3.036264054083171e-06, | |
| "loss": 0.4205, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 6.440066814422607, | |
| "learning_rate": 2.944256052444287e-06, | |
| "loss": 0.3659, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 4.709935665130615, | |
| "learning_rate": 2.852248050805403e-06, | |
| "loss": 0.405, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.5476343631744385, | |
| "learning_rate": 2.760240049166519e-06, | |
| "loss": 0.3739, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7525, | |
| "eval_f1": 0.56, | |
| "eval_loss": 0.43321692943573, | |
| "eval_precision": 0.6176470588235294, | |
| "eval_recall": 0.5121951219512195, | |
| "eval_runtime": 1.5223, | |
| "eval_samples_per_second": 262.766, | |
| "eval_steps_per_second": 16.423, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 4.812537670135498, | |
| "learning_rate": 2.668232047527635e-06, | |
| "loss": 0.3667, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 4.738429546356201, | |
| "learning_rate": 2.5762240458887513e-06, | |
| "loss": 0.4433, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.3, | |
| "grad_norm": 3.0295910835266113, | |
| "learning_rate": 2.484216044249867e-06, | |
| "loss": 0.4128, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 5.22220516204834, | |
| "learning_rate": 2.3922080426109834e-06, | |
| "loss": 0.4121, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 2.487245559692383, | |
| "learning_rate": 2.3002000409720992e-06, | |
| "loss": 0.3238, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 2.538316488265991, | |
| "learning_rate": 2.208192039333215e-06, | |
| "loss": 0.3581, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.7, | |
| "grad_norm": 5.147064685821533, | |
| "learning_rate": 2.1161840376943313e-06, | |
| "loss": 0.4136, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 7.231971263885498, | |
| "learning_rate": 2.024176036055447e-06, | |
| "loss": 0.3613, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.9, | |
| "grad_norm": 9.553585052490234, | |
| "learning_rate": 1.9321680344165634e-06, | |
| "loss": 0.3903, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 1.889197826385498, | |
| "learning_rate": 1.8401600327776794e-06, | |
| "loss": 0.3095, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.815, | |
| "eval_f1": 0.7109375, | |
| "eval_loss": 0.38830721378326416, | |
| "eval_precision": 0.6842105263157895, | |
| "eval_recall": 0.7398373983739838, | |
| "eval_runtime": 1.5185, | |
| "eval_samples_per_second": 263.415, | |
| "eval_steps_per_second": 16.463, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "grad_norm": 4.298902988433838, | |
| "learning_rate": 1.7481520311387955e-06, | |
| "loss": 0.3714, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 2.782442092895508, | |
| "learning_rate": 1.6561440294999113e-06, | |
| "loss": 0.2964, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 3.3, | |
| "grad_norm": 3.685288906097412, | |
| "learning_rate": 1.5641360278610275e-06, | |
| "loss": 0.2926, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 2.439354658126831, | |
| "learning_rate": 1.4721280262221436e-06, | |
| "loss": 0.3621, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "grad_norm": 8.328632354736328, | |
| "learning_rate": 1.3801200245832594e-06, | |
| "loss": 0.3418, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 3.328589916229248, | |
| "learning_rate": 1.2881120229443757e-06, | |
| "loss": 0.3387, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 3.7, | |
| "grad_norm": 4.276352405548096, | |
| "learning_rate": 1.1961040213054917e-06, | |
| "loss": 0.3769, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 3.431298017501831, | |
| "learning_rate": 1.1040960196666075e-06, | |
| "loss": 0.3528, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "grad_norm": 5.400251865386963, | |
| "learning_rate": 1.0120880180277236e-06, | |
| "loss": 0.2987, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 9.403071403503418, | |
| "learning_rate": 9.200800163888397e-07, | |
| "loss": 0.3237, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.84, | |
| "eval_f1": 0.746031746031746, | |
| "eval_loss": 0.3718077838420868, | |
| "eval_precision": 0.7286821705426356, | |
| "eval_recall": 0.7642276422764228, | |
| "eval_runtime": 1.5144, | |
| "eval_samples_per_second": 264.129, | |
| "eval_steps_per_second": 16.508, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.1, | |
| "grad_norm": 5.902601718902588, | |
| "learning_rate": 8.280720147499556e-07, | |
| "loss": 0.323, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "grad_norm": 7.535499095916748, | |
| "learning_rate": 7.360640131110718e-07, | |
| "loss": 0.324, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 4.3, | |
| "grad_norm": 6.525690078735352, | |
| "learning_rate": 6.440560114721878e-07, | |
| "loss": 0.3355, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 4.518765926361084, | |
| "learning_rate": 5.520480098333038e-07, | |
| "loss": 0.2986, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "grad_norm": 5.2387309074401855, | |
| "learning_rate": 4.6004000819441985e-07, | |
| "loss": 0.3432, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "grad_norm": 3.4332995414733887, | |
| "learning_rate": 3.680320065555359e-07, | |
| "loss": 0.2803, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 4.7, | |
| "grad_norm": 3.9733219146728516, | |
| "learning_rate": 2.760240049166519e-07, | |
| "loss": 0.2769, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 5.9507246017456055, | |
| "learning_rate": 1.8401600327776795e-07, | |
| "loss": 0.407, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "grad_norm": 2.992692232131958, | |
| "learning_rate": 9.200800163888397e-08, | |
| "loss": 0.2828, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 8.591716766357422, | |
| "learning_rate": 0.0, | |
| "loss": 0.3407, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8375, | |
| "eval_f1": 0.7470817120622568, | |
| "eval_loss": 0.3640846610069275, | |
| "eval_precision": 0.7164179104477612, | |
| "eval_recall": 0.7804878048780488, | |
| "eval_runtime": 1.5548, | |
| "eval_samples_per_second": 257.275, | |
| "eval_steps_per_second": 16.08, | |
| "step": 500 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1059076852254720.0, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": { | |
| "_wandb": {}, | |
| "assignments": {}, | |
| "learning_rate": 4.6004000819441984e-06, | |
| "metric": "eval/loss", | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 16, | |
| "seed": 29 | |
| } | |
| } | |