| { | |
| "best_metric": 0.6991409248766222, | |
| "best_model_checkpoint": "vit-base-patch16-224-in21k-finetuned-hongrui_mammogram_v_1/checkpoint-1710", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 1710, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05847953216374269, | |
| "grad_norm": 1.5846091508865356, | |
| "learning_rate": 2.9239766081871343e-06, | |
| "loss": 1.3844, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.11695906432748537, | |
| "grad_norm": 1.26529061794281, | |
| "learning_rate": 5.8479532163742686e-06, | |
| "loss": 1.3401, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.17543859649122806, | |
| "grad_norm": 1.0668294429779053, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 1.2478, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.23391812865497075, | |
| "grad_norm": 0.7071924805641174, | |
| "learning_rate": 1.1695906432748537e-05, | |
| "loss": 1.1693, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.29239766081871343, | |
| "grad_norm": 0.7040536403656006, | |
| "learning_rate": 1.4619883040935673e-05, | |
| "loss": 1.0874, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.3508771929824561, | |
| "grad_norm": 0.5056448578834534, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 1.0551, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.4093567251461988, | |
| "grad_norm": 0.6976526975631714, | |
| "learning_rate": 2.046783625730994e-05, | |
| "loss": 1.0349, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.4678362573099415, | |
| "grad_norm": 0.6909885406494141, | |
| "learning_rate": 2.3391812865497074e-05, | |
| "loss": 0.9862, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.5263157894736842, | |
| "grad_norm": 1.2905045747756958, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.9897, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.5847953216374269, | |
| "grad_norm": 1.0667847394943237, | |
| "learning_rate": 2.9239766081871346e-05, | |
| "loss": 0.9316, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.6432748538011696, | |
| "grad_norm": 0.7333235144615173, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.9224, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.7017543859649122, | |
| "grad_norm": 0.8830112814903259, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 0.9127, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.7602339181286549, | |
| "grad_norm": 0.7408013343811035, | |
| "learning_rate": 3.8011695906432746e-05, | |
| "loss": 0.9349, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.8187134502923976, | |
| "grad_norm": 0.6464580297470093, | |
| "learning_rate": 4.093567251461988e-05, | |
| "loss": 0.9046, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.8771929824561403, | |
| "grad_norm": 0.9568632245063782, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.8981, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.935672514619883, | |
| "grad_norm": 0.755339503288269, | |
| "learning_rate": 4.678362573099415e-05, | |
| "loss": 0.9012, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.9941520467836257, | |
| "grad_norm": 0.8921021819114685, | |
| "learning_rate": 4.970760233918128e-05, | |
| "loss": 0.8576, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6677938219703893, | |
| "eval_f1": 0.6066566978834349, | |
| "eval_loss": 0.8430724143981934, | |
| "eval_precision": 0.7751300124185229, | |
| "eval_recall": 0.6677938219703893, | |
| "eval_runtime": 78.2201, | |
| "eval_samples_per_second": 139.887, | |
| "eval_steps_per_second": 2.186, | |
| "step": 171 | |
| }, | |
| { | |
| "epoch": 1.0526315789473684, | |
| "grad_norm": 0.7332887053489685, | |
| "learning_rate": 4.970760233918128e-05, | |
| "loss": 0.8527, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.1111111111111112, | |
| "grad_norm": 0.776443362236023, | |
| "learning_rate": 4.938271604938271e-05, | |
| "loss": 0.8561, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.1695906432748537, | |
| "grad_norm": 0.7149679660797119, | |
| "learning_rate": 4.9057829759584143e-05, | |
| "loss": 0.8435, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.2280701754385965, | |
| "grad_norm": 0.8708255290985107, | |
| "learning_rate": 4.8732943469785574e-05, | |
| "loss": 0.8332, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.286549707602339, | |
| "grad_norm": 0.8141400814056396, | |
| "learning_rate": 4.8408057179987004e-05, | |
| "loss": 0.8377, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.345029239766082, | |
| "grad_norm": 1.188214659690857, | |
| "learning_rate": 4.8083170890188434e-05, | |
| "loss": 0.8456, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.4035087719298245, | |
| "grad_norm": 0.6589232087135315, | |
| "learning_rate": 4.7758284600389865e-05, | |
| "loss": 0.8387, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.4619883040935673, | |
| "grad_norm": 0.9304301142692566, | |
| "learning_rate": 4.7433398310591295e-05, | |
| "loss": 0.8521, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.52046783625731, | |
| "grad_norm": 0.8288267254829407, | |
| "learning_rate": 4.7108512020792725e-05, | |
| "loss": 0.8453, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.5789473684210527, | |
| "grad_norm": 0.8100181221961975, | |
| "learning_rate": 4.678362573099415e-05, | |
| "loss": 0.8398, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.6374269005847952, | |
| "grad_norm": 1.2519994974136353, | |
| "learning_rate": 4.645873944119558e-05, | |
| "loss": 0.835, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.695906432748538, | |
| "grad_norm": 0.868988573551178, | |
| "learning_rate": 4.613385315139701e-05, | |
| "loss": 0.8, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.7543859649122808, | |
| "grad_norm": 0.7534909844398499, | |
| "learning_rate": 4.580896686159844e-05, | |
| "loss": 0.8103, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 1.8128654970760234, | |
| "grad_norm": 0.9954193234443665, | |
| "learning_rate": 4.548408057179987e-05, | |
| "loss": 0.8065, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 1.871345029239766, | |
| "grad_norm": 0.6459550857543945, | |
| "learning_rate": 4.51591942820013e-05, | |
| "loss": 0.8196, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 1.9298245614035088, | |
| "grad_norm": 0.9698415398597717, | |
| "learning_rate": 4.483430799220273e-05, | |
| "loss": 0.8464, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 1.9883040935672516, | |
| "grad_norm": 0.9862537980079651, | |
| "learning_rate": 4.450942170240416e-05, | |
| "loss": 0.8297, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6791263023213306, | |
| "eval_f1": 0.6182249859592962, | |
| "eval_loss": 0.796485424041748, | |
| "eval_precision": 0.6757921489303887, | |
| "eval_recall": 0.6791263023213306, | |
| "eval_runtime": 78.3952, | |
| "eval_samples_per_second": 139.575, | |
| "eval_steps_per_second": 2.181, | |
| "step": 342 | |
| }, | |
| { | |
| "epoch": 2.046783625730994, | |
| "grad_norm": 0.7553840279579163, | |
| "learning_rate": 4.418453541260559e-05, | |
| "loss": 0.8262, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.1052631578947367, | |
| "grad_norm": 0.8275452852249146, | |
| "learning_rate": 4.3859649122807014e-05, | |
| "loss": 0.809, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.1637426900584797, | |
| "grad_norm": 0.6450644731521606, | |
| "learning_rate": 4.3534762833008445e-05, | |
| "loss": 0.7939, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.2222222222222223, | |
| "grad_norm": 0.6809207201004028, | |
| "learning_rate": 4.3209876543209875e-05, | |
| "loss": 0.8157, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 2.280701754385965, | |
| "grad_norm": 0.9741197228431702, | |
| "learning_rate": 4.2884990253411305e-05, | |
| "loss": 0.8126, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 2.3391812865497075, | |
| "grad_norm": 0.6929029822349548, | |
| "learning_rate": 4.2560103963612735e-05, | |
| "loss": 0.8155, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 2.39766081871345, | |
| "grad_norm": 0.836373507976532, | |
| "learning_rate": 4.2235217673814166e-05, | |
| "loss": 0.8145, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 2.456140350877193, | |
| "grad_norm": 0.7719972729682922, | |
| "learning_rate": 4.1910331384015596e-05, | |
| "loss": 0.7963, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 2.5146198830409356, | |
| "grad_norm": 0.5631088018417358, | |
| "learning_rate": 4.1585445094217026e-05, | |
| "loss": 0.7839, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 2.573099415204678, | |
| "grad_norm": 0.7167489528656006, | |
| "learning_rate": 4.1260558804418457e-05, | |
| "loss": 0.7837, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 2.6315789473684212, | |
| "grad_norm": 0.8713414072990417, | |
| "learning_rate": 4.093567251461988e-05, | |
| "loss": 0.8046, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 2.690058479532164, | |
| "grad_norm": 1.0731910467147827, | |
| "learning_rate": 4.061078622482131e-05, | |
| "loss": 0.7813, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 2.7485380116959064, | |
| "grad_norm": 0.6702953577041626, | |
| "learning_rate": 4.028589993502274e-05, | |
| "loss": 0.8076, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 2.807017543859649, | |
| "grad_norm": 0.6061651706695557, | |
| "learning_rate": 3.996101364522417e-05, | |
| "loss": 0.786, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 2.8654970760233915, | |
| "grad_norm": 0.9429291486740112, | |
| "learning_rate": 3.96361273554256e-05, | |
| "loss": 0.817, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 2.9239766081871346, | |
| "grad_norm": 0.7229118347167969, | |
| "learning_rate": 3.931124106562703e-05, | |
| "loss": 0.8003, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 2.982456140350877, | |
| "grad_norm": 0.702900230884552, | |
| "learning_rate": 3.898635477582846e-05, | |
| "loss": 0.8303, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6842441966733687, | |
| "eval_f1": 0.6360238176585905, | |
| "eval_loss": 0.7872248888015747, | |
| "eval_precision": 0.6704334160874683, | |
| "eval_recall": 0.6842441966733687, | |
| "eval_runtime": 77.468, | |
| "eval_samples_per_second": 141.245, | |
| "eval_steps_per_second": 2.207, | |
| "step": 513 | |
| }, | |
| { | |
| "epoch": 3.0409356725146197, | |
| "grad_norm": 0.6098015308380127, | |
| "learning_rate": 3.866146848602989e-05, | |
| "loss": 0.7717, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 3.0994152046783627, | |
| "grad_norm": 0.7494928240776062, | |
| "learning_rate": 3.8336582196231315e-05, | |
| "loss": 0.7807, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 3.1578947368421053, | |
| "grad_norm": 0.8039376735687256, | |
| "learning_rate": 3.8011695906432746e-05, | |
| "loss": 0.79, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 3.216374269005848, | |
| "grad_norm": 0.687044084072113, | |
| "learning_rate": 3.7686809616634176e-05, | |
| "loss": 0.8004, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 3.2748538011695905, | |
| "grad_norm": 0.9368821978569031, | |
| "learning_rate": 3.7361923326835606e-05, | |
| "loss": 0.8002, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 1.0392132997512817, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.8042, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 3.391812865497076, | |
| "grad_norm": 0.5632928013801575, | |
| "learning_rate": 3.671215074723847e-05, | |
| "loss": 0.7746, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 3.4502923976608186, | |
| "grad_norm": 0.6281052231788635, | |
| "learning_rate": 3.63872644574399e-05, | |
| "loss": 0.7691, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 3.5087719298245617, | |
| "grad_norm": 0.6072255969047546, | |
| "learning_rate": 3.606237816764133e-05, | |
| "loss": 0.7552, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 3.5672514619883042, | |
| "grad_norm": 0.6243124604225159, | |
| "learning_rate": 3.573749187784276e-05, | |
| "loss": 0.7905, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 3.625730994152047, | |
| "grad_norm": 0.8322011828422546, | |
| "learning_rate": 3.541260558804418e-05, | |
| "loss": 0.7772, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 3.6842105263157894, | |
| "grad_norm": 0.9370966553688049, | |
| "learning_rate": 3.508771929824561e-05, | |
| "loss": 0.7683, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 3.742690058479532, | |
| "grad_norm": 0.6632276177406311, | |
| "learning_rate": 3.476283300844704e-05, | |
| "loss": 0.7853, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 3.801169590643275, | |
| "grad_norm": 0.678115963935852, | |
| "learning_rate": 3.443794671864847e-05, | |
| "loss": 0.7691, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 3.8596491228070176, | |
| "grad_norm": 0.6135697960853577, | |
| "learning_rate": 3.41130604288499e-05, | |
| "loss": 0.7778, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 3.91812865497076, | |
| "grad_norm": 0.6042221188545227, | |
| "learning_rate": 3.378817413905133e-05, | |
| "loss": 0.7983, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 3.976608187134503, | |
| "grad_norm": 0.8058825135231018, | |
| "learning_rate": 3.346328784925276e-05, | |
| "loss": 0.7814, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6843355876439408, | |
| "eval_f1": 0.659727957333552, | |
| "eval_loss": 0.7717081308364868, | |
| "eval_precision": 0.6601378633948723, | |
| "eval_recall": 0.6843355876439408, | |
| "eval_runtime": 78.1449, | |
| "eval_samples_per_second": 140.022, | |
| "eval_steps_per_second": 2.188, | |
| "step": 684 | |
| }, | |
| { | |
| "epoch": 4.035087719298246, | |
| "grad_norm": 0.966820478439331, | |
| "learning_rate": 3.313840155945419e-05, | |
| "loss": 0.7736, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 4.093567251461988, | |
| "grad_norm": 1.1176632642745972, | |
| "learning_rate": 3.281351526965562e-05, | |
| "loss": 0.7529, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 4.152046783625731, | |
| "grad_norm": 0.6780201196670532, | |
| "learning_rate": 3.248862897985705e-05, | |
| "loss": 0.7722, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 4.2105263157894735, | |
| "grad_norm": 0.8257865905761719, | |
| "learning_rate": 3.216374269005848e-05, | |
| "loss": 0.771, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 4.269005847953216, | |
| "grad_norm": 0.6105393767356873, | |
| "learning_rate": 3.183885640025991e-05, | |
| "loss": 0.7883, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 4.3274853801169595, | |
| "grad_norm": 0.8537980914115906, | |
| "learning_rate": 3.151397011046134e-05, | |
| "loss": 0.7698, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 4.385964912280702, | |
| "grad_norm": 0.8124959468841553, | |
| "learning_rate": 3.118908382066277e-05, | |
| "loss": 0.7737, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 4.444444444444445, | |
| "grad_norm": 0.8992810845375061, | |
| "learning_rate": 3.08641975308642e-05, | |
| "loss": 0.7809, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 4.502923976608187, | |
| "grad_norm": 0.6706241965293884, | |
| "learning_rate": 3.053931124106563e-05, | |
| "loss": 0.7629, | |
| "step": 770 | |
| }, | |
| { | |
| "epoch": 4.56140350877193, | |
| "grad_norm": 0.8107186555862427, | |
| "learning_rate": 3.0214424951267055e-05, | |
| "loss": 0.7683, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 4.619883040935672, | |
| "grad_norm": 0.7054151296615601, | |
| "learning_rate": 2.9889538661468486e-05, | |
| "loss": 0.7367, | |
| "step": 790 | |
| }, | |
| { | |
| "epoch": 4.678362573099415, | |
| "grad_norm": 0.7284806966781616, | |
| "learning_rate": 2.9564652371669916e-05, | |
| "loss": 0.7476, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 4.7368421052631575, | |
| "grad_norm": 1.1839812994003296, | |
| "learning_rate": 2.9239766081871346e-05, | |
| "loss": 0.7565, | |
| "step": 810 | |
| }, | |
| { | |
| "epoch": 4.7953216374269, | |
| "grad_norm": 0.7781530618667603, | |
| "learning_rate": 2.8914879792072773e-05, | |
| "loss": 0.7737, | |
| "step": 820 | |
| }, | |
| { | |
| "epoch": 4.853801169590644, | |
| "grad_norm": 0.7338679432868958, | |
| "learning_rate": 2.8589993502274203e-05, | |
| "loss": 0.7519, | |
| "step": 830 | |
| }, | |
| { | |
| "epoch": 4.912280701754386, | |
| "grad_norm": 1.015286922454834, | |
| "learning_rate": 2.8265107212475634e-05, | |
| "loss": 0.8023, | |
| "step": 840 | |
| }, | |
| { | |
| "epoch": 4.970760233918129, | |
| "grad_norm": 0.8456217050552368, | |
| "learning_rate": 2.7940220922677064e-05, | |
| "loss": 0.7768, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6905501736428441, | |
| "eval_f1": 0.6543647035652124, | |
| "eval_loss": 0.7693981528282166, | |
| "eval_precision": 0.6774590515406133, | |
| "eval_recall": 0.6905501736428441, | |
| "eval_runtime": 77.1215, | |
| "eval_samples_per_second": 141.88, | |
| "eval_steps_per_second": 2.217, | |
| "step": 855 | |
| }, | |
| { | |
| "epoch": 5.029239766081871, | |
| "grad_norm": 0.624717116355896, | |
| "learning_rate": 2.761533463287849e-05, | |
| "loss": 0.7482, | |
| "step": 860 | |
| }, | |
| { | |
| "epoch": 5.087719298245614, | |
| "grad_norm": 0.752734363079071, | |
| "learning_rate": 2.729044834307992e-05, | |
| "loss": 0.7259, | |
| "step": 870 | |
| }, | |
| { | |
| "epoch": 5.146198830409356, | |
| "grad_norm": 0.6503344178199768, | |
| "learning_rate": 2.696556205328135e-05, | |
| "loss": 0.7488, | |
| "step": 880 | |
| }, | |
| { | |
| "epoch": 5.204678362573099, | |
| "grad_norm": 0.8620956540107727, | |
| "learning_rate": 2.664067576348278e-05, | |
| "loss": 0.7704, | |
| "step": 890 | |
| }, | |
| { | |
| "epoch": 5.2631578947368425, | |
| "grad_norm": 0.6938666701316833, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 0.76, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 5.321637426900585, | |
| "grad_norm": 0.8206263184547424, | |
| "learning_rate": 2.599090318388564e-05, | |
| "loss": 0.7685, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 5.380116959064328, | |
| "grad_norm": 0.8919401168823242, | |
| "learning_rate": 2.566601689408707e-05, | |
| "loss": 0.7673, | |
| "step": 920 | |
| }, | |
| { | |
| "epoch": 5.43859649122807, | |
| "grad_norm": 0.9412862062454224, | |
| "learning_rate": 2.53411306042885e-05, | |
| "loss": 0.7403, | |
| "step": 930 | |
| }, | |
| { | |
| "epoch": 5.497076023391813, | |
| "grad_norm": 1.1093353033065796, | |
| "learning_rate": 2.501624431448993e-05, | |
| "loss": 0.743, | |
| "step": 940 | |
| }, | |
| { | |
| "epoch": 5.555555555555555, | |
| "grad_norm": 0.6838064193725586, | |
| "learning_rate": 2.4691358024691357e-05, | |
| "loss": 0.7641, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 5.614035087719298, | |
| "grad_norm": 0.7546567320823669, | |
| "learning_rate": 2.4366471734892787e-05, | |
| "loss": 0.7648, | |
| "step": 960 | |
| }, | |
| { | |
| "epoch": 5.6725146198830405, | |
| "grad_norm": 0.7909204363822937, | |
| "learning_rate": 2.4041585445094217e-05, | |
| "loss": 0.7872, | |
| "step": 970 | |
| }, | |
| { | |
| "epoch": 5.730994152046784, | |
| "grad_norm": 0.6969336867332458, | |
| "learning_rate": 2.3716699155295647e-05, | |
| "loss": 0.7638, | |
| "step": 980 | |
| }, | |
| { | |
| "epoch": 5.7894736842105265, | |
| "grad_norm": 0.7838913202285767, | |
| "learning_rate": 2.3391812865497074e-05, | |
| "loss": 0.747, | |
| "step": 990 | |
| }, | |
| { | |
| "epoch": 5.847953216374269, | |
| "grad_norm": 0.8347417712211609, | |
| "learning_rate": 2.3066926575698505e-05, | |
| "loss": 0.7749, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 5.906432748538012, | |
| "grad_norm": 0.9646545052528381, | |
| "learning_rate": 2.2742040285899935e-05, | |
| "loss": 0.737, | |
| "step": 1010 | |
| }, | |
| { | |
| "epoch": 5.964912280701754, | |
| "grad_norm": 0.6134990453720093, | |
| "learning_rate": 2.2417153996101365e-05, | |
| "loss": 0.7415, | |
| "step": 1020 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6962164138183148, | |
| "eval_f1": 0.671796652878038, | |
| "eval_loss": 0.7572136521339417, | |
| "eval_precision": 0.6763761787994358, | |
| "eval_recall": 0.6962164138183148, | |
| "eval_runtime": 77.9563, | |
| "eval_samples_per_second": 140.361, | |
| "eval_steps_per_second": 2.194, | |
| "step": 1026 | |
| }, | |
| { | |
| "epoch": 6.023391812865497, | |
| "grad_norm": 0.7049497961997986, | |
| "learning_rate": 2.2092267706302795e-05, | |
| "loss": 0.7598, | |
| "step": 1030 | |
| }, | |
| { | |
| "epoch": 6.081871345029239, | |
| "grad_norm": 0.9780289530754089, | |
| "learning_rate": 2.1767381416504222e-05, | |
| "loss": 0.7472, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 6.140350877192983, | |
| "grad_norm": 0.7058891654014587, | |
| "learning_rate": 2.1442495126705653e-05, | |
| "loss": 0.742, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 6.1988304093567255, | |
| "grad_norm": 0.8734349012374878, | |
| "learning_rate": 2.1117608836907083e-05, | |
| "loss": 0.7581, | |
| "step": 1060 | |
| }, | |
| { | |
| "epoch": 6.257309941520468, | |
| "grad_norm": 0.8839743733406067, | |
| "learning_rate": 2.0792722547108513e-05, | |
| "loss": 0.7516, | |
| "step": 1070 | |
| }, | |
| { | |
| "epoch": 6.315789473684211, | |
| "grad_norm": 0.6963735222816467, | |
| "learning_rate": 2.046783625730994e-05, | |
| "loss": 0.7412, | |
| "step": 1080 | |
| }, | |
| { | |
| "epoch": 6.374269005847953, | |
| "grad_norm": 0.9337784647941589, | |
| "learning_rate": 2.014294996751137e-05, | |
| "loss": 0.7402, | |
| "step": 1090 | |
| }, | |
| { | |
| "epoch": 6.432748538011696, | |
| "grad_norm": 0.6648013591766357, | |
| "learning_rate": 1.98180636777128e-05, | |
| "loss": 0.7513, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 6.491228070175438, | |
| "grad_norm": 1.072342872619629, | |
| "learning_rate": 1.949317738791423e-05, | |
| "loss": 0.7406, | |
| "step": 1110 | |
| }, | |
| { | |
| "epoch": 6.549707602339181, | |
| "grad_norm": 1.0100135803222656, | |
| "learning_rate": 1.9168291098115658e-05, | |
| "loss": 0.7643, | |
| "step": 1120 | |
| }, | |
| { | |
| "epoch": 6.6081871345029235, | |
| "grad_norm": 0.7687884569168091, | |
| "learning_rate": 1.8843404808317088e-05, | |
| "loss": 0.7404, | |
| "step": 1130 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 0.750688374042511, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.7391, | |
| "step": 1140 | |
| }, | |
| { | |
| "epoch": 6.7251461988304095, | |
| "grad_norm": 0.7010438442230225, | |
| "learning_rate": 1.819363222871995e-05, | |
| "loss": 0.7534, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 6.783625730994152, | |
| "grad_norm": 1.0012060403823853, | |
| "learning_rate": 1.786874593892138e-05, | |
| "loss": 0.7485, | |
| "step": 1160 | |
| }, | |
| { | |
| "epoch": 6.842105263157895, | |
| "grad_norm": 0.8860548734664917, | |
| "learning_rate": 1.7543859649122806e-05, | |
| "loss": 0.7606, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 6.900584795321637, | |
| "grad_norm": 0.969633936882019, | |
| "learning_rate": 1.7218973359324236e-05, | |
| "loss": 0.7592, | |
| "step": 1180 | |
| }, | |
| { | |
| "epoch": 6.95906432748538, | |
| "grad_norm": 0.8473331928253174, | |
| "learning_rate": 1.6894087069525666e-05, | |
| "loss": 0.7351, | |
| "step": 1190 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.692195211113142, | |
| "eval_f1": 0.6568623393542826, | |
| "eval_loss": 0.754936695098877, | |
| "eval_precision": 0.6648205901494669, | |
| "eval_recall": 0.692195211113142, | |
| "eval_runtime": 77.9138, | |
| "eval_samples_per_second": 140.437, | |
| "eval_steps_per_second": 2.195, | |
| "step": 1197 | |
| }, | |
| { | |
| "epoch": 7.017543859649122, | |
| "grad_norm": 0.7504809498786926, | |
| "learning_rate": 1.6569200779727097e-05, | |
| "loss": 0.7102, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 7.076023391812866, | |
| "grad_norm": 1.306260585784912, | |
| "learning_rate": 1.6244314489928523e-05, | |
| "loss": 0.7574, | |
| "step": 1210 | |
| }, | |
| { | |
| "epoch": 7.1345029239766085, | |
| "grad_norm": 0.7214799523353577, | |
| "learning_rate": 1.5919428200129954e-05, | |
| "loss": 0.7355, | |
| "step": 1220 | |
| }, | |
| { | |
| "epoch": 7.192982456140351, | |
| "grad_norm": 0.8254335522651672, | |
| "learning_rate": 1.5594541910331384e-05, | |
| "loss": 0.758, | |
| "step": 1230 | |
| }, | |
| { | |
| "epoch": 7.251461988304094, | |
| "grad_norm": 0.8644353747367859, | |
| "learning_rate": 1.5269655620532814e-05, | |
| "loss": 0.7131, | |
| "step": 1240 | |
| }, | |
| { | |
| "epoch": 7.309941520467836, | |
| "grad_norm": 0.7876085638999939, | |
| "learning_rate": 1.4944769330734243e-05, | |
| "loss": 0.7437, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 7.368421052631579, | |
| "grad_norm": 0.904586136341095, | |
| "learning_rate": 1.4619883040935673e-05, | |
| "loss": 0.7464, | |
| "step": 1260 | |
| }, | |
| { | |
| "epoch": 7.426900584795321, | |
| "grad_norm": 1.033260464668274, | |
| "learning_rate": 1.4294996751137102e-05, | |
| "loss": 0.7226, | |
| "step": 1270 | |
| }, | |
| { | |
| "epoch": 7.485380116959064, | |
| "grad_norm": 0.9057112336158752, | |
| "learning_rate": 1.3970110461338532e-05, | |
| "loss": 0.7425, | |
| "step": 1280 | |
| }, | |
| { | |
| "epoch": 7.543859649122807, | |
| "grad_norm": 0.8631776571273804, | |
| "learning_rate": 1.364522417153996e-05, | |
| "loss": 0.7178, | |
| "step": 1290 | |
| }, | |
| { | |
| "epoch": 7.60233918128655, | |
| "grad_norm": 0.8566320538520813, | |
| "learning_rate": 1.332033788174139e-05, | |
| "loss": 0.7151, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 7.6608187134502925, | |
| "grad_norm": 1.056127905845642, | |
| "learning_rate": 1.299545159194282e-05, | |
| "loss": 0.7574, | |
| "step": 1310 | |
| }, | |
| { | |
| "epoch": 7.719298245614035, | |
| "grad_norm": 1.0582066774368286, | |
| "learning_rate": 1.267056530214425e-05, | |
| "loss": 0.7462, | |
| "step": 1320 | |
| }, | |
| { | |
| "epoch": 7.777777777777778, | |
| "grad_norm": 1.0808275938034058, | |
| "learning_rate": 1.2345679012345678e-05, | |
| "loss": 0.7181, | |
| "step": 1330 | |
| }, | |
| { | |
| "epoch": 7.83625730994152, | |
| "grad_norm": 0.8452061414718628, | |
| "learning_rate": 1.2020792722547109e-05, | |
| "loss": 0.7686, | |
| "step": 1340 | |
| }, | |
| { | |
| "epoch": 7.894736842105263, | |
| "grad_norm": 0.7253689765930176, | |
| "learning_rate": 1.1695906432748537e-05, | |
| "loss": 0.7174, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 7.953216374269006, | |
| "grad_norm": 0.9176128506660461, | |
| "learning_rate": 1.1371020142949967e-05, | |
| "loss": 0.7197, | |
| "step": 1360 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6985925790531895, | |
| "eval_f1": 0.6855055863067254, | |
| "eval_loss": 0.7478834390640259, | |
| "eval_precision": 0.6925926647987316, | |
| "eval_recall": 0.6985925790531895, | |
| "eval_runtime": 77.8555, | |
| "eval_samples_per_second": 140.542, | |
| "eval_steps_per_second": 2.196, | |
| "step": 1368 | |
| }, | |
| { | |
| "epoch": 8.011695906432749, | |
| "grad_norm": 0.7897553443908691, | |
| "learning_rate": 1.1046133853151398e-05, | |
| "loss": 0.755, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 8.070175438596491, | |
| "grad_norm": 0.7324469685554504, | |
| "learning_rate": 1.0721247563352826e-05, | |
| "loss": 0.7243, | |
| "step": 1380 | |
| }, | |
| { | |
| "epoch": 8.128654970760234, | |
| "grad_norm": 0.7983306646347046, | |
| "learning_rate": 1.0396361273554257e-05, | |
| "loss": 0.7294, | |
| "step": 1390 | |
| }, | |
| { | |
| "epoch": 8.187134502923977, | |
| "grad_norm": 0.9110460877418518, | |
| "learning_rate": 1.0071474983755685e-05, | |
| "loss": 0.7027, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 8.24561403508772, | |
| "grad_norm": 0.9574342966079712, | |
| "learning_rate": 9.746588693957115e-06, | |
| "loss": 0.7131, | |
| "step": 1410 | |
| }, | |
| { | |
| "epoch": 8.304093567251462, | |
| "grad_norm": 0.7169631719589233, | |
| "learning_rate": 9.421702404158544e-06, | |
| "loss": 0.7365, | |
| "step": 1420 | |
| }, | |
| { | |
| "epoch": 8.362573099415204, | |
| "grad_norm": 0.9551491737365723, | |
| "learning_rate": 9.096816114359974e-06, | |
| "loss": 0.7313, | |
| "step": 1430 | |
| }, | |
| { | |
| "epoch": 8.421052631578947, | |
| "grad_norm": 1.159575343132019, | |
| "learning_rate": 8.771929824561403e-06, | |
| "loss": 0.7438, | |
| "step": 1440 | |
| }, | |
| { | |
| "epoch": 8.47953216374269, | |
| "grad_norm": 0.8166360259056091, | |
| "learning_rate": 8.447043534762833e-06, | |
| "loss": 0.7355, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 8.538011695906432, | |
| "grad_norm": 0.8369165062904358, | |
| "learning_rate": 8.122157244964262e-06, | |
| "loss": 0.7183, | |
| "step": 1460 | |
| }, | |
| { | |
| "epoch": 8.596491228070175, | |
| "grad_norm": 0.7923627495765686, | |
| "learning_rate": 7.797270955165692e-06, | |
| "loss": 0.711, | |
| "step": 1470 | |
| }, | |
| { | |
| "epoch": 8.654970760233919, | |
| "grad_norm": 0.7623910903930664, | |
| "learning_rate": 7.4723846653671214e-06, | |
| "loss": 0.7377, | |
| "step": 1480 | |
| }, | |
| { | |
| "epoch": 8.713450292397662, | |
| "grad_norm": 1.088745355606079, | |
| "learning_rate": 7.147498375568551e-06, | |
| "loss": 0.7199, | |
| "step": 1490 | |
| }, | |
| { | |
| "epoch": 8.771929824561404, | |
| "grad_norm": 0.8672430515289307, | |
| "learning_rate": 6.82261208576998e-06, | |
| "loss": 0.7234, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 8.830409356725147, | |
| "grad_norm": 0.77957683801651, | |
| "learning_rate": 6.49772579597141e-06, | |
| "loss": 0.7348, | |
| "step": 1510 | |
| }, | |
| { | |
| "epoch": 8.88888888888889, | |
| "grad_norm": 0.982523500919342, | |
| "learning_rate": 6.172839506172839e-06, | |
| "loss": 0.7348, | |
| "step": 1520 | |
| }, | |
| { | |
| "epoch": 8.947368421052632, | |
| "grad_norm": 0.8758224844932556, | |
| "learning_rate": 5.8479532163742686e-06, | |
| "loss": 0.7087, | |
| "step": 1530 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.6978614512886127, | |
| "eval_f1": 0.6697434161663234, | |
| "eval_loss": 0.744518518447876, | |
| "eval_precision": 0.6792260519903555, | |
| "eval_recall": 0.6978614512886127, | |
| "eval_runtime": 77.9794, | |
| "eval_samples_per_second": 140.319, | |
| "eval_steps_per_second": 2.193, | |
| "step": 1539 | |
| }, | |
| { | |
| "epoch": 9.005847953216374, | |
| "grad_norm": 0.7864174246788025, | |
| "learning_rate": 5.523066926575699e-06, | |
| "loss": 0.713, | |
| "step": 1540 | |
| }, | |
| { | |
| "epoch": 9.064327485380117, | |
| "grad_norm": 1.0881294012069702, | |
| "learning_rate": 5.198180636777128e-06, | |
| "loss": 0.7092, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 9.12280701754386, | |
| "grad_norm": 1.0221022367477417, | |
| "learning_rate": 4.873294346978558e-06, | |
| "loss": 0.7463, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 9.181286549707602, | |
| "grad_norm": 0.8976357579231262, | |
| "learning_rate": 4.548408057179987e-06, | |
| "loss": 0.7392, | |
| "step": 1570 | |
| }, | |
| { | |
| "epoch": 9.239766081871345, | |
| "grad_norm": 0.8547308444976807, | |
| "learning_rate": 4.2235217673814166e-06, | |
| "loss": 0.7275, | |
| "step": 1580 | |
| }, | |
| { | |
| "epoch": 9.298245614035087, | |
| "grad_norm": 0.9341883063316345, | |
| "learning_rate": 3.898635477582846e-06, | |
| "loss": 0.7183, | |
| "step": 1590 | |
| }, | |
| { | |
| "epoch": 9.35672514619883, | |
| "grad_norm": 0.9447769522666931, | |
| "learning_rate": 3.5737491877842754e-06, | |
| "loss": 0.6966, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 9.415204678362572, | |
| "grad_norm": 1.0918306112289429, | |
| "learning_rate": 3.248862897985705e-06, | |
| "loss": 0.7161, | |
| "step": 1610 | |
| }, | |
| { | |
| "epoch": 9.473684210526315, | |
| "grad_norm": 0.9330850839614868, | |
| "learning_rate": 2.9239766081871343e-06, | |
| "loss": 0.7093, | |
| "step": 1620 | |
| }, | |
| { | |
| "epoch": 9.53216374269006, | |
| "grad_norm": 0.8420782685279846, | |
| "learning_rate": 2.599090318388564e-06, | |
| "loss": 0.7259, | |
| "step": 1630 | |
| }, | |
| { | |
| "epoch": 9.590643274853802, | |
| "grad_norm": 0.9159696698188782, | |
| "learning_rate": 2.2742040285899936e-06, | |
| "loss": 0.7265, | |
| "step": 1640 | |
| }, | |
| { | |
| "epoch": 9.649122807017545, | |
| "grad_norm": 1.0164194107055664, | |
| "learning_rate": 1.949317738791423e-06, | |
| "loss": 0.7458, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 9.707602339181287, | |
| "grad_norm": 0.8425694704055786, | |
| "learning_rate": 1.6244314489928524e-06, | |
| "loss": 0.7149, | |
| "step": 1660 | |
| }, | |
| { | |
| "epoch": 9.76608187134503, | |
| "grad_norm": 1.1017402410507202, | |
| "learning_rate": 1.299545159194282e-06, | |
| "loss": 0.7101, | |
| "step": 1670 | |
| }, | |
| { | |
| "epoch": 9.824561403508772, | |
| "grad_norm": 0.8220164179801941, | |
| "learning_rate": 9.746588693957115e-07, | |
| "loss": 0.7174, | |
| "step": 1680 | |
| }, | |
| { | |
| "epoch": 9.883040935672515, | |
| "grad_norm": 0.7968518733978271, | |
| "learning_rate": 6.49772579597141e-07, | |
| "loss": 0.7236, | |
| "step": 1690 | |
| }, | |
| { | |
| "epoch": 9.941520467836257, | |
| "grad_norm": 0.7491603493690491, | |
| "learning_rate": 3.248862897985705e-07, | |
| "loss": 0.6978, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.8862149119377136, | |
| "learning_rate": 0.0, | |
| "loss": 0.6977, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6991409248766222, | |
| "eval_f1": 0.6766571542539821, | |
| "eval_loss": 0.7419390082359314, | |
| "eval_precision": 0.6830062110035815, | |
| "eval_recall": 0.6991409248766222, | |
| "eval_runtime": 78.4178, | |
| "eval_samples_per_second": 139.535, | |
| "eval_steps_per_second": 2.181, | |
| "step": 1710 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 1710, | |
| "total_flos": 3.3914202248568177e+19, | |
| "train_loss": 0.790359598014787, | |
| "train_runtime": 7828.8109, | |
| "train_samples_per_second": 55.901, | |
| "train_steps_per_second": 0.218 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 1710, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "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": 3.3914202248568177e+19, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |