| { | |
| "best_metric": 0.8346071667451987, | |
| "best_model_checkpoint": "test-klue/ynat/run-3/checkpoint-1500", | |
| "epoch": 2.626970227670753, | |
| "eval_steps": 50, | |
| "global_step": 1500, | |
| "is_hyper_param_search": true, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 3.851823303957499e-06, | |
| "loss": 1.934, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_f1": 0.01705140497086806, | |
| "eval_loss": 2.002474546432495, | |
| "eval_runtime": 12.8723, | |
| "eval_samples_per_second": 707.486, | |
| "eval_steps_per_second": 1.398, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 3.783163352015298e-06, | |
| "loss": 1.9174, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_f1": 0.16773144497786774, | |
| "eval_loss": 1.9946870803833008, | |
| "eval_runtime": 13.4529, | |
| "eval_samples_per_second": 676.953, | |
| "eval_steps_per_second": 1.338, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 3.7145034000730966e-06, | |
| "loss": 1.6984, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "eval_f1": 0.579749057690188, | |
| "eval_loss": 1.6373496055603027, | |
| "eval_runtime": 13.0114, | |
| "eval_samples_per_second": 699.924, | |
| "eval_steps_per_second": 1.383, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 3.645843448130895e-06, | |
| "loss": 1.2585, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_f1": 0.6734776928120875, | |
| "eval_loss": 1.2733124494552612, | |
| "eval_runtime": 12.6224, | |
| "eval_samples_per_second": 721.493, | |
| "eval_steps_per_second": 1.426, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 3.5771834961886936e-06, | |
| "loss": 0.9487, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_f1": 0.6814953813845488, | |
| "eval_loss": 1.1104496717453003, | |
| "eval_runtime": 12.492, | |
| "eval_samples_per_second": 729.025, | |
| "eval_steps_per_second": 1.441, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 3.5085235442464923e-06, | |
| "loss": 0.8072, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_f1": 0.6717305786202503, | |
| "eval_loss": 1.0609983205795288, | |
| "eval_runtime": 12.6873, | |
| "eval_samples_per_second": 717.806, | |
| "eval_steps_per_second": 1.419, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 3.439863592304291e-06, | |
| "loss": 0.7549, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_f1": 0.7228793656517608, | |
| "eval_loss": 0.9494264721870422, | |
| "eval_runtime": 12.8373, | |
| "eval_samples_per_second": 709.416, | |
| "eval_steps_per_second": 1.402, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 3.3712036403620893e-06, | |
| "loss": 0.6644, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "eval_f1": 0.7719549711033966, | |
| "eval_loss": 0.8139678239822388, | |
| "eval_runtime": 12.9124, | |
| "eval_samples_per_second": 705.29, | |
| "eval_steps_per_second": 1.394, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "learning_rate": 3.302543688419888e-06, | |
| "loss": 0.6306, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "eval_f1": 0.7738311952589042, | |
| "eval_loss": 0.7844408750534058, | |
| "eval_runtime": 12.8236, | |
| "eval_samples_per_second": 710.174, | |
| "eval_steps_per_second": 1.404, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 3.2338837364776867e-06, | |
| "loss": 0.6093, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_f1": 0.7850534531405503, | |
| "eval_loss": 0.7450836300849915, | |
| "eval_runtime": 12.7168, | |
| "eval_samples_per_second": 716.14, | |
| "eval_steps_per_second": 1.415, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 3.1652237845354854e-06, | |
| "loss": 0.599, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "eval_f1": 0.7932118091019561, | |
| "eval_loss": 0.7213732004165649, | |
| "eval_runtime": 12.9294, | |
| "eval_samples_per_second": 704.366, | |
| "eval_steps_per_second": 1.392, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 3.0965638325932837e-06, | |
| "loss": 0.5234, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_f1": 0.8216123467232329, | |
| "eval_loss": 0.6262369751930237, | |
| "eval_runtime": 13.578, | |
| "eval_samples_per_second": 670.72, | |
| "eval_steps_per_second": 1.326, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 3.0279038806510824e-06, | |
| "loss": 0.4465, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_f1": 0.7867065499958213, | |
| "eval_loss": 0.7198110818862915, | |
| "eval_runtime": 13.2261, | |
| "eval_samples_per_second": 688.561, | |
| "eval_steps_per_second": 1.361, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 2.959243928708881e-06, | |
| "loss": 0.4521, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "eval_f1": 0.8063113996725187, | |
| "eval_loss": 0.6468276381492615, | |
| "eval_runtime": 12.8551, | |
| "eval_samples_per_second": 708.435, | |
| "eval_steps_per_second": 1.4, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 2.89058397676668e-06, | |
| "loss": 0.4339, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "eval_f1": 0.8166775515876931, | |
| "eval_loss": 0.6203234195709229, | |
| "eval_runtime": 12.669, | |
| "eval_samples_per_second": 718.842, | |
| "eval_steps_per_second": 1.421, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 2.821924024824478e-06, | |
| "loss": 0.4757, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_f1": 0.8161608316711845, | |
| "eval_loss": 0.6249090433120728, | |
| "eval_runtime": 12.9976, | |
| "eval_samples_per_second": 700.668, | |
| "eval_steps_per_second": 1.385, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "learning_rate": 2.753264072882277e-06, | |
| "loss": 0.4594, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "eval_f1": 0.8290833266193817, | |
| "eval_loss": 0.5703582167625427, | |
| "eval_runtime": 12.8799, | |
| "eval_samples_per_second": 707.072, | |
| "eval_steps_per_second": 1.398, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 2.6846041209400755e-06, | |
| "loss": 0.4032, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "eval_f1": 0.8215012797906703, | |
| "eval_loss": 0.5891036987304688, | |
| "eval_runtime": 12.8777, | |
| "eval_samples_per_second": 707.193, | |
| "eval_steps_per_second": 1.398, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "learning_rate": 2.615944168997874e-06, | |
| "loss": 0.4204, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "eval_f1": 0.8327085312817779, | |
| "eval_loss": 0.5532290935516357, | |
| "eval_runtime": 12.7986, | |
| "eval_samples_per_second": 711.563, | |
| "eval_steps_per_second": 1.406, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 2.5472842170556725e-06, | |
| "loss": 0.4386, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_f1": 0.8286389311967884, | |
| "eval_loss": 0.5586702227592468, | |
| "eval_runtime": 12.7254, | |
| "eval_samples_per_second": 715.653, | |
| "eval_steps_per_second": 1.414, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "learning_rate": 2.478624265113471e-06, | |
| "loss": 0.4258, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "eval_f1": 0.8194650098039734, | |
| "eval_loss": 0.5945160388946533, | |
| "eval_runtime": 12.8508, | |
| "eval_samples_per_second": 708.674, | |
| "eval_steps_per_second": 1.401, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 2.4099643131712695e-06, | |
| "loss": 0.4544, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "eval_f1": 0.8274559023611527, | |
| "eval_loss": 0.5760587453842163, | |
| "eval_runtime": 13.4335, | |
| "eval_samples_per_second": 677.933, | |
| "eval_steps_per_second": 1.34, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "learning_rate": 2.341304361229068e-06, | |
| "loss": 0.418, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "eval_f1": 0.8304623515332191, | |
| "eval_loss": 0.5600340962409973, | |
| "eval_runtime": 13.2414, | |
| "eval_samples_per_second": 687.766, | |
| "eval_steps_per_second": 1.359, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "learning_rate": 2.2726444092868665e-06, | |
| "loss": 0.3918, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "eval_f1": 0.8230072950822092, | |
| "eval_loss": 0.5960156321525574, | |
| "eval_runtime": 12.9264, | |
| "eval_samples_per_second": 704.529, | |
| "eval_steps_per_second": 1.393, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 2.203984457344665e-06, | |
| "loss": 0.4001, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "eval_f1": 0.8370136201480073, | |
| "eval_loss": 0.5412853360176086, | |
| "eval_runtime": 12.8464, | |
| "eval_samples_per_second": 708.917, | |
| "eval_steps_per_second": 1.401, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 2.135324505402464e-06, | |
| "loss": 0.3369, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "eval_f1": 0.8144549401231569, | |
| "eval_loss": 0.6154327392578125, | |
| "eval_runtime": 12.8518, | |
| "eval_samples_per_second": 708.616, | |
| "eval_steps_per_second": 1.401, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "learning_rate": 2.0666645534602626e-06, | |
| "loss": 0.3639, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.36, | |
| "eval_f1": 0.8218463781690494, | |
| "eval_loss": 0.6043773889541626, | |
| "eval_runtime": 12.8354, | |
| "eval_samples_per_second": 709.522, | |
| "eval_steps_per_second": 1.402, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "learning_rate": 1.998004601518061e-06, | |
| "loss": 0.305, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.45, | |
| "eval_f1": 0.8330006702059005, | |
| "eval_loss": 0.5544472336769104, | |
| "eval_runtime": 12.869, | |
| "eval_samples_per_second": 707.672, | |
| "eval_steps_per_second": 1.399, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 1.9293446495758596e-06, | |
| "loss": 0.4444, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "eval_f1": 0.8346019595913722, | |
| "eval_loss": 0.559039294719696, | |
| "eval_runtime": 12.8371, | |
| "eval_samples_per_second": 709.431, | |
| "eval_steps_per_second": 1.402, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 1.8606846976336585e-06, | |
| "loss": 0.3428, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "eval_f1": 0.8346071667451987, | |
| "eval_loss": 0.5472803115844727, | |
| "eval_runtime": 12.6738, | |
| "eval_samples_per_second": 718.567, | |
| "eval_steps_per_second": 1.42, | |
| "step": 1500 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 2855, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 122183570161200.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": { | |
| "learning_rate": 3.851823303957499e-06, | |
| "num_train_epochs": 5, | |
| "per_device_train_batch_size": 8, | |
| "seed": 30 | |
| } | |
| } | |