| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.24509803921568626, | |
| "eval_steps": 10, | |
| "global_step": 550, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.004456327985739751, | |
| "grad_norm": 0.7227747440338135, | |
| "learning_rate": 0.00019636363636363636, | |
| "loss": 0.364, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004456327985739751, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3638981878757477, | |
| "eval_runtime": 537.6403, | |
| "eval_samples_per_second": 8.348, | |
| "eval_steps_per_second": 2.087, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.008912655971479501, | |
| "grad_norm": 0.0924869254231453, | |
| "learning_rate": 0.00019272727272727274, | |
| "loss": 0.3247, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008912655971479501, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.5152587890625, | |
| "eval_runtime": 533.0372, | |
| "eval_samples_per_second": 8.42, | |
| "eval_steps_per_second": 2.105, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.013368983957219251, | |
| "grad_norm": 0.23558691143989563, | |
| "learning_rate": 0.0001890909090909091, | |
| "loss": 0.4074, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.013368983957219251, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3966202735900879, | |
| "eval_runtime": 532.3437, | |
| "eval_samples_per_second": 8.431, | |
| "eval_steps_per_second": 2.108, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.017825311942959002, | |
| "grad_norm": 0.7453870177268982, | |
| "learning_rate": 0.00018545454545454545, | |
| "loss": 0.2989, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.017825311942959002, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.33440741896629333, | |
| "eval_runtime": 532.0041, | |
| "eval_samples_per_second": 8.436, | |
| "eval_steps_per_second": 2.109, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.022281639928698752, | |
| "grad_norm": 6.811055660247803, | |
| "learning_rate": 0.00018181818181818183, | |
| "loss": 0.3399, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.022281639928698752, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.32954567670822144, | |
| "eval_runtime": 536.9612, | |
| "eval_samples_per_second": 8.358, | |
| "eval_steps_per_second": 2.09, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.026737967914438502, | |
| "grad_norm": 1.3061631917953491, | |
| "learning_rate": 0.0001781818181818182, | |
| "loss": 0.3437, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.026737967914438502, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3265528976917267, | |
| "eval_runtime": 533.3224, | |
| "eval_samples_per_second": 8.415, | |
| "eval_steps_per_second": 2.104, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.031194295900178252, | |
| "grad_norm": 1.7616266012191772, | |
| "learning_rate": 0.00017454545454545454, | |
| "loss": 0.3749, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.031194295900178252, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3122697174549103, | |
| "eval_runtime": 530.7411, | |
| "eval_samples_per_second": 8.456, | |
| "eval_steps_per_second": 2.114, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.035650623885918005, | |
| "grad_norm": 1.1275932788848877, | |
| "learning_rate": 0.0001709090909090909, | |
| "loss": 0.4267, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.035650623885918005, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.23432780802249908, | |
| "eval_runtime": 533.9058, | |
| "eval_samples_per_second": 8.406, | |
| "eval_steps_per_second": 2.101, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.040106951871657755, | |
| "grad_norm": 1.387789249420166, | |
| "learning_rate": 0.00016727272727272728, | |
| "loss": 0.4221, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.040106951871657755, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.21733854711055756, | |
| "eval_runtime": 533.5436, | |
| "eval_samples_per_second": 8.412, | |
| "eval_steps_per_second": 2.103, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.044563279857397504, | |
| "grad_norm": 9.95677375793457, | |
| "learning_rate": 0.00016363636363636366, | |
| "loss": 0.2292, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.044563279857397504, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3393175005912781, | |
| "eval_runtime": 537.6972, | |
| "eval_samples_per_second": 8.347, | |
| "eval_steps_per_second": 2.087, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.049019607843137254, | |
| "grad_norm": 1.0864927768707275, | |
| "learning_rate": 0.00016, | |
| "loss": 0.4339, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.049019607843137254, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.28164002299308777, | |
| "eval_runtime": 531.492, | |
| "eval_samples_per_second": 8.444, | |
| "eval_steps_per_second": 2.111, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.053475935828877004, | |
| "grad_norm": 1.7070204019546509, | |
| "learning_rate": 0.00015636363636363637, | |
| "loss": 0.4143, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.053475935828877004, | |
| "eval_accuracy": 0.8963903784751892, | |
| "eval_loss": 0.20752401649951935, | |
| "eval_runtime": 538.1828, | |
| "eval_samples_per_second": 8.339, | |
| "eval_steps_per_second": 2.085, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.057932263814616754, | |
| "grad_norm": 0.08961692452430725, | |
| "learning_rate": 0.00015272727272727275, | |
| "loss": 0.2757, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.057932263814616754, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.1715579330921173, | |
| "eval_runtime": 534.1189, | |
| "eval_samples_per_second": 8.403, | |
| "eval_steps_per_second": 2.101, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.062388591800356503, | |
| "grad_norm": 1.3324838876724243, | |
| "learning_rate": 0.0001490909090909091, | |
| "loss": 0.1588, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.062388591800356503, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.149879589676857, | |
| "eval_runtime": 530.7418, | |
| "eval_samples_per_second": 8.456, | |
| "eval_steps_per_second": 2.114, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.06684491978609626, | |
| "grad_norm": 2.0198991298675537, | |
| "learning_rate": 0.00014545454545454546, | |
| "loss": 0.4453, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.06684491978609626, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.2379792034626007, | |
| "eval_runtime": 530.3365, | |
| "eval_samples_per_second": 8.463, | |
| "eval_steps_per_second": 2.116, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.07130124777183601, | |
| "grad_norm": 0.3087630867958069, | |
| "learning_rate": 0.00014181818181818184, | |
| "loss": 0.2505, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.07130124777183601, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.3300795257091522, | |
| "eval_runtime": 529.0534, | |
| "eval_samples_per_second": 8.483, | |
| "eval_steps_per_second": 2.121, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.07575757575757576, | |
| "grad_norm": 2.6436047554016113, | |
| "learning_rate": 0.0001381818181818182, | |
| "loss": 0.2654, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.07575757575757576, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.22237823903560638, | |
| "eval_runtime": 536.9554, | |
| "eval_samples_per_second": 8.358, | |
| "eval_steps_per_second": 2.09, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.08021390374331551, | |
| "grad_norm": 0.9066676497459412, | |
| "learning_rate": 0.00013454545454545455, | |
| "loss": 0.3018, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.08021390374331551, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.2531951069831848, | |
| "eval_runtime": 532.2386, | |
| "eval_samples_per_second": 8.432, | |
| "eval_steps_per_second": 2.108, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.08467023172905526, | |
| "grad_norm": 0.10262551158666611, | |
| "learning_rate": 0.00013090909090909093, | |
| "loss": 0.2325, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.08467023172905526, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.22432851791381836, | |
| "eval_runtime": 531.5973, | |
| "eval_samples_per_second": 8.442, | |
| "eval_steps_per_second": 2.111, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.08912655971479501, | |
| "grad_norm": 0.20511843264102936, | |
| "learning_rate": 0.00012727272727272728, | |
| "loss": 0.2641, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.08912655971479501, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.14137917757034302, | |
| "eval_runtime": 528.9957, | |
| "eval_samples_per_second": 8.484, | |
| "eval_steps_per_second": 2.121, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.09358288770053476, | |
| "grad_norm": 0.9824792742729187, | |
| "learning_rate": 0.00012363636363636364, | |
| "loss": 0.1497, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.09358288770053476, | |
| "eval_accuracy": 0.89683598279953, | |
| "eval_loss": 0.12781496345996857, | |
| "eval_runtime": 532.3045, | |
| "eval_samples_per_second": 8.431, | |
| "eval_steps_per_second": 2.108, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.09803921568627451, | |
| "grad_norm": 0.29042425751686096, | |
| "learning_rate": 0.00012, | |
| "loss": 0.1536, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.09803921568627451, | |
| "eval_accuracy": 0.9380570650100708, | |
| "eval_loss": 0.14640431106090546, | |
| "eval_runtime": 535.3538, | |
| "eval_samples_per_second": 8.383, | |
| "eval_steps_per_second": 2.096, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.10249554367201426, | |
| "grad_norm": 0.5411375761032104, | |
| "learning_rate": 0.00011636363636363636, | |
| "loss": 0.1801, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.10249554367201426, | |
| "eval_accuracy": 0.9358288645744324, | |
| "eval_loss": 0.1414562165737152, | |
| "eval_runtime": 533.0939, | |
| "eval_samples_per_second": 8.419, | |
| "eval_steps_per_second": 2.105, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.10695187165775401, | |
| "grad_norm": 5.630716323852539, | |
| "learning_rate": 0.00011272727272727272, | |
| "loss": 0.1344, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.10695187165775401, | |
| "eval_accuracy": 0.9844028353691101, | |
| "eval_loss": 0.08324988931417465, | |
| "eval_runtime": 536.1639, | |
| "eval_samples_per_second": 8.371, | |
| "eval_steps_per_second": 2.093, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.11140819964349376, | |
| "grad_norm": 0.27143993973731995, | |
| "learning_rate": 0.00010909090909090909, | |
| "loss": 0.1722, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.11140819964349376, | |
| "eval_accuracy": 0.9861853718757629, | |
| "eval_loss": 0.060878392308950424, | |
| "eval_runtime": 533.7467, | |
| "eval_samples_per_second": 8.408, | |
| "eval_steps_per_second": 2.102, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.11586452762923351, | |
| "grad_norm": 0.6671731472015381, | |
| "learning_rate": 0.00010545454545454545, | |
| "loss": 0.0684, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.11586452762923351, | |
| "eval_accuracy": 0.8997326493263245, | |
| "eval_loss": 0.19059808552265167, | |
| "eval_runtime": 537.3352, | |
| "eval_samples_per_second": 8.352, | |
| "eval_steps_per_second": 2.088, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.12032085561497326, | |
| "grad_norm": 0.007157750893384218, | |
| "learning_rate": 0.00010181818181818181, | |
| "loss": 0.2185, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.12032085561497326, | |
| "eval_accuracy": 0.9703654050827026, | |
| "eval_loss": 0.07430911809206009, | |
| "eval_runtime": 534.3595, | |
| "eval_samples_per_second": 8.399, | |
| "eval_steps_per_second": 2.1, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.12477718360071301, | |
| "grad_norm": 0.10907144099473953, | |
| "learning_rate": 9.818181818181818e-05, | |
| "loss": 0.0562, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.12477718360071301, | |
| "eval_accuracy": 0.9839572310447693, | |
| "eval_loss": 0.046000149101018906, | |
| "eval_runtime": 532.5534, | |
| "eval_samples_per_second": 8.427, | |
| "eval_steps_per_second": 2.107, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.12923351158645277, | |
| "grad_norm": 0.004431632813066244, | |
| "learning_rate": 9.454545454545455e-05, | |
| "loss": 0.007, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.12923351158645277, | |
| "eval_accuracy": 0.9895276427268982, | |
| "eval_loss": 0.03365661948919296, | |
| "eval_runtime": 539.0883, | |
| "eval_samples_per_second": 8.325, | |
| "eval_steps_per_second": 2.081, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.13368983957219252, | |
| "grad_norm": 0.14239944517612457, | |
| "learning_rate": 9.090909090909092e-05, | |
| "loss": 0.1054, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13368983957219252, | |
| "eval_accuracy": 0.9890819787979126, | |
| "eval_loss": 0.03497695177793503, | |
| "eval_runtime": 530.7114, | |
| "eval_samples_per_second": 8.457, | |
| "eval_steps_per_second": 2.114, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.13814616755793227, | |
| "grad_norm": 0.19534932076931, | |
| "learning_rate": 8.727272727272727e-05, | |
| "loss": 0.0074, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.13814616755793227, | |
| "eval_accuracy": 0.9815062284469604, | |
| "eval_loss": 0.055771518498659134, | |
| "eval_runtime": 540.1438, | |
| "eval_samples_per_second": 8.309, | |
| "eval_steps_per_second": 2.077, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.14260249554367202, | |
| "grad_norm": 0.14106573164463043, | |
| "learning_rate": 8.363636363636364e-05, | |
| "loss": 0.0069, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.14260249554367202, | |
| "eval_accuracy": 0.9752673506736755, | |
| "eval_loss": 0.07411307096481323, | |
| "eval_runtime": 530.0836, | |
| "eval_samples_per_second": 8.467, | |
| "eval_steps_per_second": 2.117, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.14705882352941177, | |
| "grad_norm": 0.04756563529372215, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0881, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.14705882352941177, | |
| "eval_accuracy": 0.9884135723114014, | |
| "eval_loss": 0.03824571892619133, | |
| "eval_runtime": 529.2936, | |
| "eval_samples_per_second": 8.479, | |
| "eval_steps_per_second": 2.12, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.15151515151515152, | |
| "grad_norm": 0.008182384073734283, | |
| "learning_rate": 7.636363636363637e-05, | |
| "loss": 0.047, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.15151515151515152, | |
| "eval_accuracy": 0.9779411554336548, | |
| "eval_loss": 0.07092902809381485, | |
| "eval_runtime": 530.9951, | |
| "eval_samples_per_second": 8.452, | |
| "eval_steps_per_second": 2.113, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.15597147950089127, | |
| "grad_norm": 6.055085182189941, | |
| "learning_rate": 7.272727272727273e-05, | |
| "loss": 0.1135, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.15597147950089127, | |
| "eval_accuracy": 0.9935383200645447, | |
| "eval_loss": 0.02391628548502922, | |
| "eval_runtime": 531.8853, | |
| "eval_samples_per_second": 8.438, | |
| "eval_steps_per_second": 2.109, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.16042780748663102, | |
| "grad_norm": 2.0424458980560303, | |
| "learning_rate": 6.90909090909091e-05, | |
| "loss": 0.1708, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.16042780748663102, | |
| "eval_accuracy": 0.9844028353691101, | |
| "eval_loss": 0.0482899434864521, | |
| "eval_runtime": 538.9554, | |
| "eval_samples_per_second": 8.327, | |
| "eval_steps_per_second": 2.082, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.16488413547237077, | |
| "grad_norm": 0.008309995755553246, | |
| "learning_rate": 6.545454545454546e-05, | |
| "loss": 0.0053, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.16488413547237077, | |
| "eval_accuracy": 0.9977718591690063, | |
| "eval_loss": 0.011557623744010925, | |
| "eval_runtime": 536.7458, | |
| "eval_samples_per_second": 8.361, | |
| "eval_steps_per_second": 2.09, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.16934046345811052, | |
| "grad_norm": 0.0617559477686882, | |
| "learning_rate": 6.181818181818182e-05, | |
| "loss": 0.0257, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.16934046345811052, | |
| "eval_accuracy": 0.9979946613311768, | |
| "eval_loss": 0.011457420885562897, | |
| "eval_runtime": 538.0907, | |
| "eval_samples_per_second": 8.341, | |
| "eval_steps_per_second": 2.085, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.17379679144385027, | |
| "grad_norm": 0.1585519164800644, | |
| "learning_rate": 5.818181818181818e-05, | |
| "loss": 0.0978, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.17379679144385027, | |
| "eval_accuracy": 0.9986631274223328, | |
| "eval_loss": 0.009089282713830471, | |
| "eval_runtime": 539.9652, | |
| "eval_samples_per_second": 8.312, | |
| "eval_steps_per_second": 2.078, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.17825311942959002, | |
| "grad_norm": 0.13566212356090546, | |
| "learning_rate": 5.4545454545454546e-05, | |
| "loss": 0.0467, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.17825311942959002, | |
| "eval_accuracy": 0.9982174634933472, | |
| "eval_loss": 0.011012092232704163, | |
| "eval_runtime": 539.0037, | |
| "eval_samples_per_second": 8.326, | |
| "eval_steps_per_second": 2.082, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.18270944741532977, | |
| "grad_norm": 0.054975979030132294, | |
| "learning_rate": 5.090909090909091e-05, | |
| "loss": 0.0266, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.18270944741532977, | |
| "eval_accuracy": 0.9988859295845032, | |
| "eval_loss": 0.008680622093379498, | |
| "eval_runtime": 536.4536, | |
| "eval_samples_per_second": 8.366, | |
| "eval_steps_per_second": 2.092, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 0.18716577540106952, | |
| "grad_norm": 9.600521087646484, | |
| "learning_rate": 4.7272727272727275e-05, | |
| "loss": 0.0118, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.18716577540106952, | |
| "eval_accuracy": 0.9988859295845032, | |
| "eval_loss": 0.008196841925382614, | |
| "eval_runtime": 529.3899, | |
| "eval_samples_per_second": 8.478, | |
| "eval_steps_per_second": 2.119, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 0.19162210338680927, | |
| "grad_norm": 0.00836893916130066, | |
| "learning_rate": 4.3636363636363636e-05, | |
| "loss": 0.0689, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.19162210338680927, | |
| "eval_accuracy": 0.9984402656555176, | |
| "eval_loss": 0.009529507718980312, | |
| "eval_runtime": 537.9471, | |
| "eval_samples_per_second": 8.343, | |
| "eval_steps_per_second": 2.086, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 0.19607843137254902, | |
| "grad_norm": 0.05235498398542404, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0785, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.19607843137254902, | |
| "eval_accuracy": 0.9975489974021912, | |
| "eval_loss": 0.01092607993632555, | |
| "eval_runtime": 533.7576, | |
| "eval_samples_per_second": 8.408, | |
| "eval_steps_per_second": 2.102, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 0.20053475935828877, | |
| "grad_norm": 7.876805782318115, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.0749, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.20053475935828877, | |
| "eval_accuracy": 0.9971033930778503, | |
| "eval_loss": 0.013591339811682701, | |
| "eval_runtime": 532.153, | |
| "eval_samples_per_second": 8.434, | |
| "eval_steps_per_second": 2.108, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.20499108734402852, | |
| "grad_norm": 0.055418092757463455, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 0.0038, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.20499108734402852, | |
| "eval_accuracy": 0.9930927157402039, | |
| "eval_loss": 0.025715434923768044, | |
| "eval_runtime": 531.9415, | |
| "eval_samples_per_second": 8.437, | |
| "eval_steps_per_second": 2.109, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 0.20944741532976827, | |
| "grad_norm": 26.78069305419922, | |
| "learning_rate": 2.909090909090909e-05, | |
| "loss": 0.1342, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.20944741532976827, | |
| "eval_accuracy": 0.987522304058075, | |
| "eval_loss": 0.040662843734025955, | |
| "eval_runtime": 530.1651, | |
| "eval_samples_per_second": 8.465, | |
| "eval_steps_per_second": 2.116, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 0.21390374331550802, | |
| "grad_norm": 13.968317031860352, | |
| "learning_rate": 2.5454545454545454e-05, | |
| "loss": 0.1816, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.21390374331550802, | |
| "eval_accuracy": 0.9803921580314636, | |
| "eval_loss": 0.06171296164393425, | |
| "eval_runtime": 536.4017, | |
| "eval_samples_per_second": 8.367, | |
| "eval_steps_per_second": 2.092, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 0.21836007130124777, | |
| "grad_norm": 0.009305565617978573, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 0.0735, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.21836007130124777, | |
| "eval_accuracy": 0.9734848737716675, | |
| "eval_loss": 0.08175662159919739, | |
| "eval_runtime": 530.5285, | |
| "eval_samples_per_second": 8.459, | |
| "eval_steps_per_second": 2.115, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 0.22281639928698752, | |
| "grad_norm": 0.00870052631944418, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.0535, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.22281639928698752, | |
| "eval_accuracy": 0.9748217463493347, | |
| "eval_loss": 0.07746395468711853, | |
| "eval_runtime": 534.4791, | |
| "eval_samples_per_second": 8.397, | |
| "eval_steps_per_second": 2.099, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.22727272727272727, | |
| "grad_norm": 0.05970863625407219, | |
| "learning_rate": 1.4545454545454545e-05, | |
| "loss": 0.0088, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.22727272727272727, | |
| "eval_accuracy": 0.9846256971359253, | |
| "eval_loss": 0.04695257917046547, | |
| "eval_runtime": 532.8984, | |
| "eval_samples_per_second": 8.422, | |
| "eval_steps_per_second": 2.105, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 0.23172905525846701, | |
| "grad_norm": 11.049286842346191, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 0.0836, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.23172905525846701, | |
| "eval_accuracy": 0.9897504448890686, | |
| "eval_loss": 0.03497246652841568, | |
| "eval_runtime": 534.5562, | |
| "eval_samples_per_second": 8.396, | |
| "eval_steps_per_second": 2.099, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 0.23618538324420676, | |
| "grad_norm": 0.18157783150672913, | |
| "learning_rate": 7.272727272727272e-06, | |
| "loss": 0.0086, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.23618538324420676, | |
| "eval_accuracy": 0.9888591766357422, | |
| "eval_loss": 0.03722322732210159, | |
| "eval_runtime": 532.0239, | |
| "eval_samples_per_second": 8.436, | |
| "eval_steps_per_second": 2.109, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 0.24064171122994651, | |
| "grad_norm": 0.35108181834220886, | |
| "learning_rate": 3.636363636363636e-06, | |
| "loss": 0.0046, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.24064171122994651, | |
| "eval_accuracy": 0.9888591766357422, | |
| "eval_loss": 0.0368054136633873, | |
| "eval_runtime": 531.8153, | |
| "eval_samples_per_second": 8.439, | |
| "eval_steps_per_second": 2.11, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 0.24509803921568626, | |
| "grad_norm": 0.008579758927226067, | |
| "learning_rate": 0.0, | |
| "loss": 0.042, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.24509803921568626, | |
| "eval_accuracy": 0.9890819787979126, | |
| "eval_loss": 0.03674088791012764, | |
| "eval_runtime": 537.7421, | |
| "eval_samples_per_second": 8.346, | |
| "eval_steps_per_second": 2.087, | |
| "step": 550 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 550, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.326400520422712e+17, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |