| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 124.03846153846153, | |
| "eval_steps": 500, | |
| "global_step": 12900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9615384615384616, | |
| "grad_norm": 165.56590270996094, | |
| "learning_rate": 9.922558139534884e-05, | |
| "loss": 48.5372, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 36.333778381347656, | |
| "eval_runtime": 21.9927, | |
| "eval_samples_per_second": 2.183, | |
| "eval_steps_per_second": 0.455, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 143.93603515625, | |
| "learning_rate": 9.845116279069768e-05, | |
| "loss": 33.5327, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.735132217407227, | |
| "eval_runtime": 16.9414, | |
| "eval_samples_per_second": 2.833, | |
| "eval_steps_per_second": 0.59, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 2.8846153846153846, | |
| "grad_norm": 135.81661987304688, | |
| "learning_rate": 9.767674418604652e-05, | |
| "loss": 29.7691, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.485803604125977, | |
| "eval_runtime": 16.7561, | |
| "eval_samples_per_second": 2.865, | |
| "eval_steps_per_second": 0.597, | |
| "step": 312 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 110.2176284790039, | |
| "learning_rate": 9.690232558139534e-05, | |
| "loss": 26.3002, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.60906219482422, | |
| "eval_runtime": 16.9439, | |
| "eval_samples_per_second": 2.833, | |
| "eval_steps_per_second": 0.59, | |
| "step": 416 | |
| }, | |
| { | |
| "epoch": 4.8076923076923075, | |
| "grad_norm": 112.39118194580078, | |
| "learning_rate": 9.612790697674419e-05, | |
| "loss": 24.7501, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 27.096742630004883, | |
| "eval_runtime": 16.7521, | |
| "eval_samples_per_second": 2.865, | |
| "eval_steps_per_second": 0.597, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 5.769230769230769, | |
| "grad_norm": 86.3100357055664, | |
| "learning_rate": 9.535348837209303e-05, | |
| "loss": 23.4495, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 26.624135971069336, | |
| "eval_runtime": 16.6013, | |
| "eval_samples_per_second": 2.891, | |
| "eval_steps_per_second": 0.602, | |
| "step": 624 | |
| }, | |
| { | |
| "epoch": 6.730769230769231, | |
| "grad_norm": 58.464332580566406, | |
| "learning_rate": 9.457906976744187e-05, | |
| "loss": 23.274, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 27.154443740844727, | |
| "eval_runtime": 17.6955, | |
| "eval_samples_per_second": 2.713, | |
| "eval_steps_per_second": 0.565, | |
| "step": 728 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 42.398406982421875, | |
| "learning_rate": 9.38046511627907e-05, | |
| "loss": 21.1617, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 27.462488174438477, | |
| "eval_runtime": 16.2589, | |
| "eval_samples_per_second": 2.952, | |
| "eval_steps_per_second": 0.615, | |
| "step": 832 | |
| }, | |
| { | |
| "epoch": 8.653846153846153, | |
| "grad_norm": 69.99388122558594, | |
| "learning_rate": 9.303023255813954e-05, | |
| "loss": 20.373, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 27.574462890625, | |
| "eval_runtime": 17.0559, | |
| "eval_samples_per_second": 2.814, | |
| "eval_steps_per_second": 0.586, | |
| "step": 936 | |
| }, | |
| { | |
| "epoch": 9.615384615384615, | |
| "grad_norm": 51.642242431640625, | |
| "learning_rate": 9.225581395348837e-05, | |
| "loss": 20.4295, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 27.69423484802246, | |
| "eval_runtime": 16.3331, | |
| "eval_samples_per_second": 2.939, | |
| "eval_steps_per_second": 0.612, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 10.576923076923077, | |
| "grad_norm": 39.88796615600586, | |
| "learning_rate": 9.14813953488372e-05, | |
| "loss": 20.2526, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 27.782920837402344, | |
| "eval_runtime": 16.9662, | |
| "eval_samples_per_second": 2.829, | |
| "eval_steps_per_second": 0.589, | |
| "step": 1144 | |
| }, | |
| { | |
| "epoch": 11.538461538461538, | |
| "grad_norm": 50.71794509887695, | |
| "learning_rate": 9.070697674418606e-05, | |
| "loss": 19.2572, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 27.295957565307617, | |
| "eval_runtime": 16.7472, | |
| "eval_samples_per_second": 2.866, | |
| "eval_steps_per_second": 0.597, | |
| "step": 1248 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 45.02073669433594, | |
| "learning_rate": 8.99325581395349e-05, | |
| "loss": 19.0089, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 26.0039005279541, | |
| "eval_runtime": 16.8217, | |
| "eval_samples_per_second": 2.853, | |
| "eval_steps_per_second": 0.594, | |
| "step": 1352 | |
| }, | |
| { | |
| "epoch": 13.461538461538462, | |
| "grad_norm": 33.62112808227539, | |
| "learning_rate": 8.915813953488373e-05, | |
| "loss": 18.3621, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 26.562255859375, | |
| "eval_runtime": 16.6581, | |
| "eval_samples_per_second": 2.881, | |
| "eval_steps_per_second": 0.6, | |
| "step": 1456 | |
| }, | |
| { | |
| "epoch": 14.423076923076923, | |
| "grad_norm": 44.2991943359375, | |
| "learning_rate": 8.838372093023257e-05, | |
| "loss": 18.0517, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 26.27000617980957, | |
| "eval_runtime": 17.496, | |
| "eval_samples_per_second": 2.743, | |
| "eval_steps_per_second": 0.572, | |
| "step": 1560 | |
| }, | |
| { | |
| "epoch": 15.384615384615385, | |
| "grad_norm": 44.68825912475586, | |
| "learning_rate": 8.76093023255814e-05, | |
| "loss": 18.3139, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 27.297170639038086, | |
| "eval_runtime": 16.2819, | |
| "eval_samples_per_second": 2.948, | |
| "eval_steps_per_second": 0.614, | |
| "step": 1664 | |
| }, | |
| { | |
| "epoch": 16.346153846153847, | |
| "grad_norm": 50.15152359008789, | |
| "learning_rate": 8.683488372093023e-05, | |
| "loss": 17.6129, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 26.486886978149414, | |
| "eval_runtime": 17.6532, | |
| "eval_samples_per_second": 2.719, | |
| "eval_steps_per_second": 0.566, | |
| "step": 1768 | |
| }, | |
| { | |
| "epoch": 17.307692307692307, | |
| "grad_norm": 31.138574600219727, | |
| "learning_rate": 8.606046511627907e-05, | |
| "loss": 17.8402, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 27.761808395385742, | |
| "eval_runtime": 16.6098, | |
| "eval_samples_per_second": 2.89, | |
| "eval_steps_per_second": 0.602, | |
| "step": 1872 | |
| }, | |
| { | |
| "epoch": 18.26923076923077, | |
| "grad_norm": 45.22633361816406, | |
| "learning_rate": 8.52860465116279e-05, | |
| "loss": 16.6494, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 27.517309188842773, | |
| "eval_runtime": 16.9988, | |
| "eval_samples_per_second": 2.824, | |
| "eval_steps_per_second": 0.588, | |
| "step": 1976 | |
| }, | |
| { | |
| "epoch": 19.23076923076923, | |
| "grad_norm": 25.58769989013672, | |
| "learning_rate": 8.451162790697674e-05, | |
| "loss": 17.0833, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.124242782592773, | |
| "eval_runtime": 16.6787, | |
| "eval_samples_per_second": 2.878, | |
| "eval_steps_per_second": 0.6, | |
| "step": 2080 | |
| }, | |
| { | |
| "epoch": 20.192307692307693, | |
| "grad_norm": 30.538772583007812, | |
| "learning_rate": 8.373720930232558e-05, | |
| "loss": 16.5967, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.119504928588867, | |
| "eval_runtime": 16.9877, | |
| "eval_samples_per_second": 2.826, | |
| "eval_steps_per_second": 0.589, | |
| "step": 2184 | |
| }, | |
| { | |
| "epoch": 21.153846153846153, | |
| "grad_norm": 34.11674499511719, | |
| "learning_rate": 8.296279069767443e-05, | |
| "loss": 16.2634, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 27.036653518676758, | |
| "eval_runtime": 16.3157, | |
| "eval_samples_per_second": 2.942, | |
| "eval_steps_per_second": 0.613, | |
| "step": 2288 | |
| }, | |
| { | |
| "epoch": 22.115384615384617, | |
| "grad_norm": 23.20075798034668, | |
| "learning_rate": 8.218837209302326e-05, | |
| "loss": 16.6797, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 27.179914474487305, | |
| "eval_runtime": 17.6647, | |
| "eval_samples_per_second": 2.717, | |
| "eval_steps_per_second": 0.566, | |
| "step": 2392 | |
| }, | |
| { | |
| "epoch": 23.076923076923077, | |
| "grad_norm": 38.62594223022461, | |
| "learning_rate": 8.14139534883721e-05, | |
| "loss": 16.0344, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 26.640764236450195, | |
| "eval_runtime": 17.4121, | |
| "eval_samples_per_second": 2.757, | |
| "eval_steps_per_second": 0.574, | |
| "step": 2496 | |
| }, | |
| { | |
| "epoch": 24.03846153846154, | |
| "grad_norm": 25.629343032836914, | |
| "learning_rate": 8.063953488372093e-05, | |
| "loss": 15.8706, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "grad_norm": 108.9260482788086, | |
| "learning_rate": 7.986511627906977e-05, | |
| "loss": 15.7701, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.403966903686523, | |
| "eval_runtime": 16.6389, | |
| "eval_samples_per_second": 2.885, | |
| "eval_steps_per_second": 0.601, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 25.96153846153846, | |
| "grad_norm": 39.876651763916016, | |
| "learning_rate": 7.90906976744186e-05, | |
| "loss": 15.6061, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.06866455078125, | |
| "eval_runtime": 17.057, | |
| "eval_samples_per_second": 2.814, | |
| "eval_steps_per_second": 0.586, | |
| "step": 2704 | |
| }, | |
| { | |
| "epoch": 26.923076923076923, | |
| "grad_norm": 24.31376838684082, | |
| "learning_rate": 7.831627906976746e-05, | |
| "loss": 15.3311, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 27.17653465270996, | |
| "eval_runtime": 16.4838, | |
| "eval_samples_per_second": 2.912, | |
| "eval_steps_per_second": 0.607, | |
| "step": 2808 | |
| }, | |
| { | |
| "epoch": 27.884615384615383, | |
| "grad_norm": 21.51975440979004, | |
| "learning_rate": 7.754186046511628e-05, | |
| "loss": 15.2464, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.20502471923828, | |
| "eval_runtime": 17.0908, | |
| "eval_samples_per_second": 2.809, | |
| "eval_steps_per_second": 0.585, | |
| "step": 2912 | |
| }, | |
| { | |
| "epoch": 28.846153846153847, | |
| "grad_norm": 25.89150047302246, | |
| "learning_rate": 7.676744186046512e-05, | |
| "loss": 15.0459, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.629125595092773, | |
| "eval_runtime": 17.0746, | |
| "eval_samples_per_second": 2.811, | |
| "eval_steps_per_second": 0.586, | |
| "step": 3016 | |
| }, | |
| { | |
| "epoch": 29.807692307692307, | |
| "grad_norm": 24.885116577148438, | |
| "learning_rate": 7.599302325581396e-05, | |
| "loss": 14.7514, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 27.824087142944336, | |
| "eval_runtime": 16.6658, | |
| "eval_samples_per_second": 2.88, | |
| "eval_steps_per_second": 0.6, | |
| "step": 3120 | |
| }, | |
| { | |
| "epoch": 30.76923076923077, | |
| "grad_norm": 23.02336883544922, | |
| "learning_rate": 7.52186046511628e-05, | |
| "loss": 15.0833, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 31.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.19361686706543, | |
| "eval_runtime": 19.4392, | |
| "eval_samples_per_second": 2.469, | |
| "eval_steps_per_second": 0.514, | |
| "step": 3224 | |
| }, | |
| { | |
| "epoch": 31.73076923076923, | |
| "grad_norm": 24.152498245239258, | |
| "learning_rate": 7.444418604651163e-05, | |
| "loss": 15.0817, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.40435218811035, | |
| "eval_runtime": 16.6685, | |
| "eval_samples_per_second": 2.88, | |
| "eval_steps_per_second": 0.6, | |
| "step": 3328 | |
| }, | |
| { | |
| "epoch": 32.69230769230769, | |
| "grad_norm": 20.687313079833984, | |
| "learning_rate": 7.366976744186047e-05, | |
| "loss": 14.3201, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 33.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.370887756347656, | |
| "eval_runtime": 16.9055, | |
| "eval_samples_per_second": 2.839, | |
| "eval_steps_per_second": 0.592, | |
| "step": 3432 | |
| }, | |
| { | |
| "epoch": 33.65384615384615, | |
| "grad_norm": 18.918672561645508, | |
| "learning_rate": 7.289534883720931e-05, | |
| "loss": 14.5918, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.38982582092285, | |
| "eval_runtime": 17.1503, | |
| "eval_samples_per_second": 2.799, | |
| "eval_steps_per_second": 0.583, | |
| "step": 3536 | |
| }, | |
| { | |
| "epoch": 34.61538461538461, | |
| "grad_norm": 34.329986572265625, | |
| "learning_rate": 7.212093023255815e-05, | |
| "loss": 14.7177, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 35.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.512975692749023, | |
| "eval_runtime": 17.026, | |
| "eval_samples_per_second": 2.819, | |
| "eval_steps_per_second": 0.587, | |
| "step": 3640 | |
| }, | |
| { | |
| "epoch": 35.57692307692308, | |
| "grad_norm": 14.776827812194824, | |
| "learning_rate": 7.134651162790697e-05, | |
| "loss": 13.9919, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 27.759668350219727, | |
| "eval_runtime": 16.7517, | |
| "eval_samples_per_second": 2.865, | |
| "eval_steps_per_second": 0.597, | |
| "step": 3744 | |
| }, | |
| { | |
| "epoch": 36.53846153846154, | |
| "grad_norm": 33.82184982299805, | |
| "learning_rate": 7.057209302325581e-05, | |
| "loss": 14.2267, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 37.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.232383728027344, | |
| "eval_runtime": 16.949, | |
| "eval_samples_per_second": 2.832, | |
| "eval_steps_per_second": 0.59, | |
| "step": 3848 | |
| }, | |
| { | |
| "epoch": 37.5, | |
| "grad_norm": 71.82172393798828, | |
| "learning_rate": 6.979767441860466e-05, | |
| "loss": 13.7801, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 38.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.357421875, | |
| "eval_runtime": 16.6259, | |
| "eval_samples_per_second": 2.887, | |
| "eval_steps_per_second": 0.601, | |
| "step": 3952 | |
| }, | |
| { | |
| "epoch": 38.46153846153846, | |
| "grad_norm": 24.108745574951172, | |
| "learning_rate": 6.90232558139535e-05, | |
| "loss": 14.1839, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 39.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.8710994720459, | |
| "eval_runtime": 16.7167, | |
| "eval_samples_per_second": 2.871, | |
| "eval_steps_per_second": 0.598, | |
| "step": 4056 | |
| }, | |
| { | |
| "epoch": 39.42307692307692, | |
| "grad_norm": 24.0490665435791, | |
| "learning_rate": 6.824883720930233e-05, | |
| "loss": 13.7545, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.294702529907227, | |
| "eval_runtime": 16.5809, | |
| "eval_samples_per_second": 2.895, | |
| "eval_steps_per_second": 0.603, | |
| "step": 4160 | |
| }, | |
| { | |
| "epoch": 40.38461538461539, | |
| "grad_norm": 17.7755184173584, | |
| "learning_rate": 6.747441860465117e-05, | |
| "loss": 14.1627, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 41.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.48655128479004, | |
| "eval_runtime": 16.9288, | |
| "eval_samples_per_second": 2.835, | |
| "eval_steps_per_second": 0.591, | |
| "step": 4264 | |
| }, | |
| { | |
| "epoch": 41.34615384615385, | |
| "grad_norm": 25.07582664489746, | |
| "learning_rate": 6.670000000000001e-05, | |
| "loss": 13.5155, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 42.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.852659225463867, | |
| "eval_runtime": 16.5485, | |
| "eval_samples_per_second": 2.901, | |
| "eval_steps_per_second": 0.604, | |
| "step": 4368 | |
| }, | |
| { | |
| "epoch": 42.30769230769231, | |
| "grad_norm": 35.709686279296875, | |
| "learning_rate": 6.592558139534885e-05, | |
| "loss": 13.704, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 43.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.4291934967041, | |
| "eval_runtime": 16.8607, | |
| "eval_samples_per_second": 2.847, | |
| "eval_steps_per_second": 0.593, | |
| "step": 4472 | |
| }, | |
| { | |
| "epoch": 43.26923076923077, | |
| "grad_norm": 21.939144134521484, | |
| "learning_rate": 6.515116279069768e-05, | |
| "loss": 13.6644, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.23241424560547, | |
| "eval_runtime": 16.6868, | |
| "eval_samples_per_second": 2.877, | |
| "eval_steps_per_second": 0.599, | |
| "step": 4576 | |
| }, | |
| { | |
| "epoch": 44.23076923076923, | |
| "grad_norm": 22.5328426361084, | |
| "learning_rate": 6.437674418604652e-05, | |
| "loss": 13.2006, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 45.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.541372299194336, | |
| "eval_runtime": 16.3804, | |
| "eval_samples_per_second": 2.93, | |
| "eval_steps_per_second": 0.61, | |
| "step": 4680 | |
| }, | |
| { | |
| "epoch": 45.19230769230769, | |
| "grad_norm": 19.775171279907227, | |
| "learning_rate": 6.360232558139536e-05, | |
| "loss": 13.1545, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 46.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.698814392089844, | |
| "eval_runtime": 17.0544, | |
| "eval_samples_per_second": 2.815, | |
| "eval_steps_per_second": 0.586, | |
| "step": 4784 | |
| }, | |
| { | |
| "epoch": 46.15384615384615, | |
| "grad_norm": 23.579957962036133, | |
| "learning_rate": 6.28279069767442e-05, | |
| "loss": 13.5744, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 47.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.993318557739258, | |
| "eval_runtime": 16.712, | |
| "eval_samples_per_second": 2.872, | |
| "eval_steps_per_second": 0.598, | |
| "step": 4888 | |
| }, | |
| { | |
| "epoch": 47.11538461538461, | |
| "grad_norm": 16.147850036621094, | |
| "learning_rate": 6.205348837209302e-05, | |
| "loss": 12.8073, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 28.976961135864258, | |
| "eval_runtime": 17.0515, | |
| "eval_samples_per_second": 2.815, | |
| "eval_steps_per_second": 0.586, | |
| "step": 4992 | |
| }, | |
| { | |
| "epoch": 48.07692307692308, | |
| "grad_norm": 39.781028747558594, | |
| "learning_rate": 6.127906976744186e-05, | |
| "loss": 13.3773, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 49.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.39496421813965, | |
| "eval_runtime": 16.6402, | |
| "eval_samples_per_second": 2.885, | |
| "eval_steps_per_second": 0.601, | |
| "step": 5096 | |
| }, | |
| { | |
| "epoch": 49.03846153846154, | |
| "grad_norm": 28.625978469848633, | |
| "learning_rate": 6.0504651162790696e-05, | |
| "loss": 12.9449, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "grad_norm": 30.444610595703125, | |
| "learning_rate": 5.9730232558139533e-05, | |
| "loss": 12.9506, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 50.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.28708839416504, | |
| "eval_runtime": 16.7982, | |
| "eval_samples_per_second": 2.857, | |
| "eval_steps_per_second": 0.595, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 50.96153846153846, | |
| "grad_norm": 22.916648864746094, | |
| "learning_rate": 5.895581395348837e-05, | |
| "loss": 13.0674, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 51.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.57108497619629, | |
| "eval_runtime": 16.4065, | |
| "eval_samples_per_second": 2.926, | |
| "eval_steps_per_second": 0.61, | |
| "step": 5304 | |
| }, | |
| { | |
| "epoch": 51.92307692307692, | |
| "grad_norm": 21.012855529785156, | |
| "learning_rate": 5.818139534883721e-05, | |
| "loss": 13.1265, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.08868980407715, | |
| "eval_runtime": 16.8841, | |
| "eval_samples_per_second": 2.843, | |
| "eval_steps_per_second": 0.592, | |
| "step": 5408 | |
| }, | |
| { | |
| "epoch": 52.88461538461539, | |
| "grad_norm": 16.615829467773438, | |
| "learning_rate": 5.740697674418606e-05, | |
| "loss": 13.1392, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 53.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.843284606933594, | |
| "eval_runtime": 17.0042, | |
| "eval_samples_per_second": 2.823, | |
| "eval_steps_per_second": 0.588, | |
| "step": 5512 | |
| }, | |
| { | |
| "epoch": 53.84615384615385, | |
| "grad_norm": 17.38899040222168, | |
| "learning_rate": 5.6632558139534884e-05, | |
| "loss": 12.6108, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 54.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.643640518188477, | |
| "eval_runtime": 16.9549, | |
| "eval_samples_per_second": 2.831, | |
| "eval_steps_per_second": 0.59, | |
| "step": 5616 | |
| }, | |
| { | |
| "epoch": 54.80769230769231, | |
| "grad_norm": 20.021800994873047, | |
| "learning_rate": 5.585813953488372e-05, | |
| "loss": 12.7608, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 55.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.870615005493164, | |
| "eval_runtime": 16.4959, | |
| "eval_samples_per_second": 2.91, | |
| "eval_steps_per_second": 0.606, | |
| "step": 5720 | |
| }, | |
| { | |
| "epoch": 55.76923076923077, | |
| "grad_norm": 19.976686477661133, | |
| "learning_rate": 5.5083720930232566e-05, | |
| "loss": 12.8723, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.059600830078125, | |
| "eval_runtime": 16.9691, | |
| "eval_samples_per_second": 2.829, | |
| "eval_steps_per_second": 0.589, | |
| "step": 5824 | |
| }, | |
| { | |
| "epoch": 56.73076923076923, | |
| "grad_norm": 34.17774963378906, | |
| "learning_rate": 5.4309302325581404e-05, | |
| "loss": 12.5437, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 57.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.136695861816406, | |
| "eval_runtime": 16.1925, | |
| "eval_samples_per_second": 2.964, | |
| "eval_steps_per_second": 0.618, | |
| "step": 5928 | |
| }, | |
| { | |
| "epoch": 57.69230769230769, | |
| "grad_norm": 16.256919860839844, | |
| "learning_rate": 5.353488372093024e-05, | |
| "loss": 12.1387, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 58.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.4089298248291, | |
| "eval_runtime": 17.0149, | |
| "eval_samples_per_second": 2.821, | |
| "eval_steps_per_second": 0.588, | |
| "step": 6032 | |
| }, | |
| { | |
| "epoch": 58.65384615384615, | |
| "grad_norm": 24.035823822021484, | |
| "learning_rate": 5.276046511627908e-05, | |
| "loss": 12.948, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 59.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.537500381469727, | |
| "eval_runtime": 16.6335, | |
| "eval_samples_per_second": 2.886, | |
| "eval_steps_per_second": 0.601, | |
| "step": 6136 | |
| }, | |
| { | |
| "epoch": 59.61538461538461, | |
| "grad_norm": 18.700029373168945, | |
| "learning_rate": 5.1986046511627916e-05, | |
| "loss": 12.2869, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.38269805908203, | |
| "eval_runtime": 16.6776, | |
| "eval_samples_per_second": 2.878, | |
| "eval_steps_per_second": 0.6, | |
| "step": 6240 | |
| }, | |
| { | |
| "epoch": 60.57692307692308, | |
| "grad_norm": 34.31249237060547, | |
| "learning_rate": 5.121162790697675e-05, | |
| "loss": 12.7717, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 61.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.639678955078125, | |
| "eval_runtime": 16.6901, | |
| "eval_samples_per_second": 2.876, | |
| "eval_steps_per_second": 0.599, | |
| "step": 6344 | |
| }, | |
| { | |
| "epoch": 61.53846153846154, | |
| "grad_norm": 26.939144134521484, | |
| "learning_rate": 5.0437209302325585e-05, | |
| "loss": 12.4924, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 62.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.700525283813477, | |
| "eval_runtime": 16.6341, | |
| "eval_samples_per_second": 2.886, | |
| "eval_steps_per_second": 0.601, | |
| "step": 6448 | |
| }, | |
| { | |
| "epoch": 62.5, | |
| "grad_norm": 13.92809009552002, | |
| "learning_rate": 4.966279069767442e-05, | |
| "loss": 12.3031, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 63.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 29.986501693725586, | |
| "eval_runtime": 16.2627, | |
| "eval_samples_per_second": 2.952, | |
| "eval_steps_per_second": 0.615, | |
| "step": 6552 | |
| }, | |
| { | |
| "epoch": 63.46153846153846, | |
| "grad_norm": 53.49395751953125, | |
| "learning_rate": 4.888837209302326e-05, | |
| "loss": 12.5575, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.06968116760254, | |
| "eval_runtime": 16.634, | |
| "eval_samples_per_second": 2.886, | |
| "eval_steps_per_second": 0.601, | |
| "step": 6656 | |
| }, | |
| { | |
| "epoch": 64.42307692307692, | |
| "grad_norm": 20.313173294067383, | |
| "learning_rate": 4.811395348837209e-05, | |
| "loss": 11.9496, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 65.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.579355239868164, | |
| "eval_runtime": 16.5822, | |
| "eval_samples_per_second": 2.895, | |
| "eval_steps_per_second": 0.603, | |
| "step": 6760 | |
| }, | |
| { | |
| "epoch": 65.38461538461539, | |
| "grad_norm": 33.59545135498047, | |
| "learning_rate": 4.733953488372093e-05, | |
| "loss": 12.0462, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 66.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.65366554260254, | |
| "eval_runtime": 16.6216, | |
| "eval_samples_per_second": 2.888, | |
| "eval_steps_per_second": 0.602, | |
| "step": 6864 | |
| }, | |
| { | |
| "epoch": 66.34615384615384, | |
| "grad_norm": 41.989036560058594, | |
| "learning_rate": 4.656511627906977e-05, | |
| "loss": 12.7167, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 67.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.74114990234375, | |
| "eval_runtime": 16.9556, | |
| "eval_samples_per_second": 2.831, | |
| "eval_steps_per_second": 0.59, | |
| "step": 6968 | |
| }, | |
| { | |
| "epoch": 67.3076923076923, | |
| "grad_norm": 39.545814514160156, | |
| "learning_rate": 4.579069767441861e-05, | |
| "loss": 11.8595, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.4969539642334, | |
| "eval_runtime": 16.5848, | |
| "eval_samples_per_second": 2.894, | |
| "eval_steps_per_second": 0.603, | |
| "step": 7072 | |
| }, | |
| { | |
| "epoch": 68.26923076923077, | |
| "grad_norm": 17.500286102294922, | |
| "learning_rate": 4.501627906976745e-05, | |
| "loss": 11.7458, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 69.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.833221435546875, | |
| "eval_runtime": 16.7855, | |
| "eval_samples_per_second": 2.86, | |
| "eval_steps_per_second": 0.596, | |
| "step": 7176 | |
| }, | |
| { | |
| "epoch": 69.23076923076923, | |
| "grad_norm": 18.128597259521484, | |
| "learning_rate": 4.4241860465116286e-05, | |
| "loss": 12.2058, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 70.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.0950813293457, | |
| "eval_runtime": 17.2742, | |
| "eval_samples_per_second": 2.779, | |
| "eval_steps_per_second": 0.579, | |
| "step": 7280 | |
| }, | |
| { | |
| "epoch": 70.1923076923077, | |
| "grad_norm": 25.363075256347656, | |
| "learning_rate": 4.3467441860465117e-05, | |
| "loss": 12.0874, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 71.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.46952438354492, | |
| "eval_runtime": 17.4937, | |
| "eval_samples_per_second": 2.744, | |
| "eval_steps_per_second": 0.572, | |
| "step": 7384 | |
| }, | |
| { | |
| "epoch": 71.15384615384616, | |
| "grad_norm": 9.088305473327637, | |
| "learning_rate": 4.2693023255813954e-05, | |
| "loss": 11.705, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.311721801757812, | |
| "eval_runtime": 16.6157, | |
| "eval_samples_per_second": 2.889, | |
| "eval_steps_per_second": 0.602, | |
| "step": 7488 | |
| }, | |
| { | |
| "epoch": 72.11538461538461, | |
| "grad_norm": 21.332366943359375, | |
| "learning_rate": 4.191860465116279e-05, | |
| "loss": 12.0, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 73.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.654035568237305, | |
| "eval_runtime": 17.001, | |
| "eval_samples_per_second": 2.823, | |
| "eval_steps_per_second": 0.588, | |
| "step": 7592 | |
| }, | |
| { | |
| "epoch": 73.07692307692308, | |
| "grad_norm": 16.92377281188965, | |
| "learning_rate": 4.114418604651163e-05, | |
| "loss": 11.9852, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 74.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 34.29496765136719, | |
| "eval_runtime": 16.6247, | |
| "eval_samples_per_second": 2.887, | |
| "eval_steps_per_second": 0.602, | |
| "step": 7696 | |
| }, | |
| { | |
| "epoch": 74.03846153846153, | |
| "grad_norm": 47.94444274902344, | |
| "learning_rate": 4.0369767441860474e-05, | |
| "loss": 11.8647, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "grad_norm": 31.79910659790039, | |
| "learning_rate": 3.959534883720931e-05, | |
| "loss": 11.7597, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 75.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.636077880859375, | |
| "eval_runtime": 16.64, | |
| "eval_samples_per_second": 2.885, | |
| "eval_steps_per_second": 0.601, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 75.96153846153847, | |
| "grad_norm": 17.918701171875, | |
| "learning_rate": 3.882093023255814e-05, | |
| "loss": 11.8713, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.10820198059082, | |
| "eval_runtime": 16.6206, | |
| "eval_samples_per_second": 2.888, | |
| "eval_steps_per_second": 0.602, | |
| "step": 7904 | |
| }, | |
| { | |
| "epoch": 76.92307692307692, | |
| "grad_norm": 34.80290985107422, | |
| "learning_rate": 3.804651162790698e-05, | |
| "loss": 11.705, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 77.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.87137794494629, | |
| "eval_runtime": 16.986, | |
| "eval_samples_per_second": 2.826, | |
| "eval_steps_per_second": 0.589, | |
| "step": 8008 | |
| }, | |
| { | |
| "epoch": 77.88461538461539, | |
| "grad_norm": 16.33148193359375, | |
| "learning_rate": 3.727209302325582e-05, | |
| "loss": 11.5474, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 78.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.029922485351562, | |
| "eval_runtime": 16.598, | |
| "eval_samples_per_second": 2.892, | |
| "eval_steps_per_second": 0.602, | |
| "step": 8112 | |
| }, | |
| { | |
| "epoch": 78.84615384615384, | |
| "grad_norm": 15.126340866088867, | |
| "learning_rate": 3.6497674418604655e-05, | |
| "loss": 11.8387, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 79.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.367172241210938, | |
| "eval_runtime": 16.6089, | |
| "eval_samples_per_second": 2.89, | |
| "eval_steps_per_second": 0.602, | |
| "step": 8216 | |
| }, | |
| { | |
| "epoch": 79.8076923076923, | |
| "grad_norm": 13.069074630737305, | |
| "learning_rate": 3.5723255813953486e-05, | |
| "loss": 11.7057, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.643484115600586, | |
| "eval_runtime": 16.4881, | |
| "eval_samples_per_second": 2.911, | |
| "eval_steps_per_second": 0.606, | |
| "step": 8320 | |
| }, | |
| { | |
| "epoch": 80.76923076923077, | |
| "grad_norm": 24.73469352722168, | |
| "learning_rate": 3.4948837209302323e-05, | |
| "loss": 11.5656, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 81.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.194046020507812, | |
| "eval_runtime": 16.6985, | |
| "eval_samples_per_second": 2.875, | |
| "eval_steps_per_second": 0.599, | |
| "step": 8424 | |
| }, | |
| { | |
| "epoch": 81.73076923076923, | |
| "grad_norm": 26.05837631225586, | |
| "learning_rate": 3.417441860465117e-05, | |
| "loss": 11.6578, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 82.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.818357467651367, | |
| "eval_runtime": 16.6321, | |
| "eval_samples_per_second": 2.886, | |
| "eval_steps_per_second": 0.601, | |
| "step": 8528 | |
| }, | |
| { | |
| "epoch": 82.6923076923077, | |
| "grad_norm": 29.041229248046875, | |
| "learning_rate": 3.3400000000000005e-05, | |
| "loss": 11.3049, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 83.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.866790771484375, | |
| "eval_runtime": 16.6756, | |
| "eval_samples_per_second": 2.878, | |
| "eval_steps_per_second": 0.6, | |
| "step": 8632 | |
| }, | |
| { | |
| "epoch": 83.65384615384616, | |
| "grad_norm": 14.160347938537598, | |
| "learning_rate": 3.262558139534884e-05, | |
| "loss": 11.5542, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.81920623779297, | |
| "eval_runtime": 16.6408, | |
| "eval_samples_per_second": 2.884, | |
| "eval_steps_per_second": 0.601, | |
| "step": 8736 | |
| }, | |
| { | |
| "epoch": 84.61538461538461, | |
| "grad_norm": 11.912644386291504, | |
| "learning_rate": 3.185116279069768e-05, | |
| "loss": 11.3942, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 85.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 30.972320556640625, | |
| "eval_runtime": 16.7795, | |
| "eval_samples_per_second": 2.861, | |
| "eval_steps_per_second": 0.596, | |
| "step": 8840 | |
| }, | |
| { | |
| "epoch": 85.57692307692308, | |
| "grad_norm": 19.492916107177734, | |
| "learning_rate": 3.107674418604651e-05, | |
| "loss": 11.6955, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 86.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.348657608032227, | |
| "eval_runtime": 16.743, | |
| "eval_samples_per_second": 2.867, | |
| "eval_steps_per_second": 0.597, | |
| "step": 8944 | |
| }, | |
| { | |
| "epoch": 86.53846153846153, | |
| "grad_norm": 14.419718742370605, | |
| "learning_rate": 3.0302325581395346e-05, | |
| "loss": 11.4862, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 87.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.045135498046875, | |
| "eval_runtime": 16.7069, | |
| "eval_samples_per_second": 2.873, | |
| "eval_steps_per_second": 0.599, | |
| "step": 9048 | |
| }, | |
| { | |
| "epoch": 87.5, | |
| "grad_norm": 24.9334659576416, | |
| "learning_rate": 2.9527906976744187e-05, | |
| "loss": 11.5867, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.976896286010742, | |
| "eval_runtime": 16.5865, | |
| "eval_samples_per_second": 2.894, | |
| "eval_steps_per_second": 0.603, | |
| "step": 9152 | |
| }, | |
| { | |
| "epoch": 88.46153846153847, | |
| "grad_norm": 10.158917427062988, | |
| "learning_rate": 2.8753488372093018e-05, | |
| "loss": 11.0975, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 89.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.972078323364258, | |
| "eval_runtime": 16.6959, | |
| "eval_samples_per_second": 2.875, | |
| "eval_steps_per_second": 0.599, | |
| "step": 9256 | |
| }, | |
| { | |
| "epoch": 89.42307692307692, | |
| "grad_norm": 11.408435821533203, | |
| "learning_rate": 2.797906976744187e-05, | |
| "loss": 11.5126, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 90.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 35.387664794921875, | |
| "eval_runtime": 16.6889, | |
| "eval_samples_per_second": 2.876, | |
| "eval_steps_per_second": 0.599, | |
| "step": 9360 | |
| }, | |
| { | |
| "epoch": 90.38461538461539, | |
| "grad_norm": 18.4831600189209, | |
| "learning_rate": 2.72046511627907e-05, | |
| "loss": 11.067, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 91.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.76138687133789, | |
| "eval_runtime": 16.2325, | |
| "eval_samples_per_second": 2.957, | |
| "eval_steps_per_second": 0.616, | |
| "step": 9464 | |
| }, | |
| { | |
| "epoch": 91.34615384615384, | |
| "grad_norm": 14.10232925415039, | |
| "learning_rate": 2.643023255813954e-05, | |
| "loss": 11.3857, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.704566955566406, | |
| "eval_runtime": 16.6368, | |
| "eval_samples_per_second": 2.885, | |
| "eval_steps_per_second": 0.601, | |
| "step": 9568 | |
| }, | |
| { | |
| "epoch": 92.3076923076923, | |
| "grad_norm": 10.36620807647705, | |
| "learning_rate": 2.565581395348837e-05, | |
| "loss": 11.5511, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 93.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.10957336425781, | |
| "eval_runtime": 16.8564, | |
| "eval_samples_per_second": 2.848, | |
| "eval_steps_per_second": 0.593, | |
| "step": 9672 | |
| }, | |
| { | |
| "epoch": 93.26923076923077, | |
| "grad_norm": 13.063793182373047, | |
| "learning_rate": 2.488139534883721e-05, | |
| "loss": 11.0961, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 94.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.8302001953125, | |
| "eval_runtime": 16.6194, | |
| "eval_samples_per_second": 2.888, | |
| "eval_steps_per_second": 0.602, | |
| "step": 9776 | |
| }, | |
| { | |
| "epoch": 94.23076923076923, | |
| "grad_norm": 47.0684814453125, | |
| "learning_rate": 2.4106976744186043e-05, | |
| "loss": 11.2935, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 95.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.66879653930664, | |
| "eval_runtime": 16.8788, | |
| "eval_samples_per_second": 2.844, | |
| "eval_steps_per_second": 0.592, | |
| "step": 9880 | |
| }, | |
| { | |
| "epoch": 95.1923076923077, | |
| "grad_norm": 16.18996238708496, | |
| "learning_rate": 2.333255813953488e-05, | |
| "loss": 11.2398, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 96.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.2806510925293, | |
| "eval_runtime": 16.4167, | |
| "eval_samples_per_second": 2.924, | |
| "eval_steps_per_second": 0.609, | |
| "step": 9984 | |
| }, | |
| { | |
| "epoch": 96.15384615384616, | |
| "grad_norm": 20.301530838012695, | |
| "learning_rate": 2.2558139534883715e-05, | |
| "loss": 11.0444, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 97.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.276641845703125, | |
| "eval_runtime": 16.7261, | |
| "eval_samples_per_second": 2.87, | |
| "eval_steps_per_second": 0.598, | |
| "step": 10088 | |
| }, | |
| { | |
| "epoch": 97.11538461538461, | |
| "grad_norm": 16.50993537902832, | |
| "learning_rate": 2.1783720930232563e-05, | |
| "loss": 11.3157, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 98.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.443729400634766, | |
| "eval_runtime": 16.7525, | |
| "eval_samples_per_second": 2.865, | |
| "eval_steps_per_second": 0.597, | |
| "step": 10192 | |
| }, | |
| { | |
| "epoch": 98.07692307692308, | |
| "grad_norm": 13.487881660461426, | |
| "learning_rate": 2.1009302325581397e-05, | |
| "loss": 11.0191, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 99.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.385108947753906, | |
| "eval_runtime": 16.6154, | |
| "eval_samples_per_second": 2.889, | |
| "eval_steps_per_second": 0.602, | |
| "step": 10296 | |
| }, | |
| { | |
| "epoch": 99.03846153846153, | |
| "grad_norm": 13.905572891235352, | |
| "learning_rate": 2.0234883720930235e-05, | |
| "loss": 11.2286, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "grad_norm": 37.709373474121094, | |
| "learning_rate": 1.946046511627907e-05, | |
| "loss": 11.1406, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 100.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.138919830322266, | |
| "eval_runtime": 16.6356, | |
| "eval_samples_per_second": 2.885, | |
| "eval_steps_per_second": 0.601, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 100.96153846153847, | |
| "grad_norm": 9.66380500793457, | |
| "learning_rate": 1.8686046511627907e-05, | |
| "loss": 11.1237, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 101.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.488643646240234, | |
| "eval_runtime": 16.5945, | |
| "eval_samples_per_second": 2.893, | |
| "eval_steps_per_second": 0.603, | |
| "step": 10504 | |
| }, | |
| { | |
| "epoch": 101.92307692307692, | |
| "grad_norm": 8.697158813476562, | |
| "learning_rate": 1.791162790697674e-05, | |
| "loss": 10.9485, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 102.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.50514602661133, | |
| "eval_runtime": 16.4431, | |
| "eval_samples_per_second": 2.919, | |
| "eval_steps_per_second": 0.608, | |
| "step": 10608 | |
| }, | |
| { | |
| "epoch": 102.88461538461539, | |
| "grad_norm": 7.563882350921631, | |
| "learning_rate": 1.713720930232558e-05, | |
| "loss": 10.9188, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 103.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.86152648925781, | |
| "eval_runtime": 16.6025, | |
| "eval_samples_per_second": 2.891, | |
| "eval_steps_per_second": 0.602, | |
| "step": 10712 | |
| }, | |
| { | |
| "epoch": 103.84615384615384, | |
| "grad_norm": 9.366110801696777, | |
| "learning_rate": 1.6362790697674413e-05, | |
| "loss": 11.3029, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 104.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.038761138916016, | |
| "eval_runtime": 16.8928, | |
| "eval_samples_per_second": 2.841, | |
| "eval_steps_per_second": 0.592, | |
| "step": 10816 | |
| }, | |
| { | |
| "epoch": 104.8076923076923, | |
| "grad_norm": 28.598913192749023, | |
| "learning_rate": 1.558837209302326e-05, | |
| "loss": 11.2023, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 105.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.492279052734375, | |
| "eval_runtime": 16.6606, | |
| "eval_samples_per_second": 2.881, | |
| "eval_steps_per_second": 0.6, | |
| "step": 10920 | |
| }, | |
| { | |
| "epoch": 105.76923076923077, | |
| "grad_norm": 23.246570587158203, | |
| "learning_rate": 1.4813953488372098e-05, | |
| "loss": 10.9634, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 106.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.32876968383789, | |
| "eval_runtime": 16.8242, | |
| "eval_samples_per_second": 2.853, | |
| "eval_steps_per_second": 0.594, | |
| "step": 11024 | |
| }, | |
| { | |
| "epoch": 106.73076923076923, | |
| "grad_norm": 12.571370124816895, | |
| "learning_rate": 1.4039534883720934e-05, | |
| "loss": 11.257, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 107.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 31.88549041748047, | |
| "eval_runtime": 16.349, | |
| "eval_samples_per_second": 2.936, | |
| "eval_steps_per_second": 0.612, | |
| "step": 11128 | |
| }, | |
| { | |
| "epoch": 107.6923076923077, | |
| "grad_norm": 26.309329986572266, | |
| "learning_rate": 1.326511627906977e-05, | |
| "loss": 11.0193, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 108.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 34.006710052490234, | |
| "eval_runtime": 17.4954, | |
| "eval_samples_per_second": 2.744, | |
| "eval_steps_per_second": 0.572, | |
| "step": 11232 | |
| }, | |
| { | |
| "epoch": 108.65384615384616, | |
| "grad_norm": 21.006258010864258, | |
| "learning_rate": 1.2490697674418606e-05, | |
| "loss": 10.6401, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 109.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.29460144042969, | |
| "eval_runtime": 16.6851, | |
| "eval_samples_per_second": 2.877, | |
| "eval_steps_per_second": 0.599, | |
| "step": 11336 | |
| }, | |
| { | |
| "epoch": 109.61538461538461, | |
| "grad_norm": 17.321447372436523, | |
| "learning_rate": 1.1716279069767442e-05, | |
| "loss": 11.0542, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 110.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 34.05351638793945, | |
| "eval_runtime": 16.6555, | |
| "eval_samples_per_second": 2.882, | |
| "eval_steps_per_second": 0.6, | |
| "step": 11440 | |
| }, | |
| { | |
| "epoch": 110.57692307692308, | |
| "grad_norm": 15.087108612060547, | |
| "learning_rate": 1.0941860465116278e-05, | |
| "loss": 10.888, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 111.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.72056198120117, | |
| "eval_runtime": 16.9072, | |
| "eval_samples_per_second": 2.839, | |
| "eval_steps_per_second": 0.591, | |
| "step": 11544 | |
| }, | |
| { | |
| "epoch": 111.53846153846153, | |
| "grad_norm": 8.586106300354004, | |
| "learning_rate": 1.0167441860465113e-05, | |
| "loss": 10.9706, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 112.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.12382888793945, | |
| "eval_runtime": 16.5804, | |
| "eval_samples_per_second": 2.895, | |
| "eval_steps_per_second": 0.603, | |
| "step": 11648 | |
| }, | |
| { | |
| "epoch": 112.5, | |
| "grad_norm": 7.436498165130615, | |
| "learning_rate": 9.39302325581396e-06, | |
| "loss": 11.0075, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 113.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.988155364990234, | |
| "eval_runtime": 16.4227, | |
| "eval_samples_per_second": 2.923, | |
| "eval_steps_per_second": 0.609, | |
| "step": 11752 | |
| }, | |
| { | |
| "epoch": 113.46153846153847, | |
| "grad_norm": 8.474443435668945, | |
| "learning_rate": 8.618604651162795e-06, | |
| "loss": 10.7895, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 114.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.79851150512695, | |
| "eval_runtime": 17.5126, | |
| "eval_samples_per_second": 2.741, | |
| "eval_steps_per_second": 0.571, | |
| "step": 11856 | |
| }, | |
| { | |
| "epoch": 114.42307692307692, | |
| "grad_norm": 17.689861297607422, | |
| "learning_rate": 7.844186046511631e-06, | |
| "loss": 10.9181, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 115.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 32.91426086425781, | |
| "eval_runtime": 17.7135, | |
| "eval_samples_per_second": 2.71, | |
| "eval_steps_per_second": 0.565, | |
| "step": 11960 | |
| }, | |
| { | |
| "epoch": 115.38461538461539, | |
| "grad_norm": 10.550300598144531, | |
| "learning_rate": 7.069767441860467e-06, | |
| "loss": 10.5938, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 116.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.07219314575195, | |
| "eval_runtime": 16.7996, | |
| "eval_samples_per_second": 2.857, | |
| "eval_steps_per_second": 0.595, | |
| "step": 12064 | |
| }, | |
| { | |
| "epoch": 116.34615384615384, | |
| "grad_norm": 10.545084953308105, | |
| "learning_rate": 6.295348837209302e-06, | |
| "loss": 10.4932, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 117.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 34.2365837097168, | |
| "eval_runtime": 17.0204, | |
| "eval_samples_per_second": 2.82, | |
| "eval_steps_per_second": 0.588, | |
| "step": 12168 | |
| }, | |
| { | |
| "epoch": 117.3076923076923, | |
| "grad_norm": 11.177499771118164, | |
| "learning_rate": 5.520930232558138e-06, | |
| "loss": 10.9761, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 118.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.88801956176758, | |
| "eval_runtime": 16.5143, | |
| "eval_samples_per_second": 2.907, | |
| "eval_steps_per_second": 0.606, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 118.26923076923077, | |
| "grad_norm": 9.421356201171875, | |
| "learning_rate": 4.746511627906974e-06, | |
| "loss": 10.6918, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 119.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 34.32889175415039, | |
| "eval_runtime": 17.2621, | |
| "eval_samples_per_second": 2.781, | |
| "eval_steps_per_second": 0.579, | |
| "step": 12376 | |
| }, | |
| { | |
| "epoch": 119.23076923076923, | |
| "grad_norm": 11.958416938781738, | |
| "learning_rate": 3.97209302325581e-06, | |
| "loss": 10.896, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 120.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.60952377319336, | |
| "eval_runtime": 17.2959, | |
| "eval_samples_per_second": 2.775, | |
| "eval_steps_per_second": 0.578, | |
| "step": 12480 | |
| }, | |
| { | |
| "epoch": 120.1923076923077, | |
| "grad_norm": 7.985867023468018, | |
| "learning_rate": 3.1976744186046562e-06, | |
| "loss": 10.6876, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 121.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.86077880859375, | |
| "eval_runtime": 17.1725, | |
| "eval_samples_per_second": 2.795, | |
| "eval_steps_per_second": 0.582, | |
| "step": 12584 | |
| }, | |
| { | |
| "epoch": 121.15384615384616, | |
| "grad_norm": 7.811230182647705, | |
| "learning_rate": 2.4232558139534926e-06, | |
| "loss": 10.5666, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 122.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.69937515258789, | |
| "eval_runtime": 16.6446, | |
| "eval_samples_per_second": 2.884, | |
| "eval_steps_per_second": 0.601, | |
| "step": 12688 | |
| }, | |
| { | |
| "epoch": 122.11538461538461, | |
| "grad_norm": 6.883728504180908, | |
| "learning_rate": 1.6488372093023285e-06, | |
| "loss": 10.8161, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 123.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.61716842651367, | |
| "eval_runtime": 17.0639, | |
| "eval_samples_per_second": 2.813, | |
| "eval_steps_per_second": 0.586, | |
| "step": 12792 | |
| }, | |
| { | |
| "epoch": 123.07692307692308, | |
| "grad_norm": 7.218296527862549, | |
| "learning_rate": 8.744186046511642e-07, | |
| "loss": 10.7195, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 124.0, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.539737701416016, | |
| "eval_runtime": 16.3513, | |
| "eval_samples_per_second": 2.936, | |
| "eval_steps_per_second": 0.612, | |
| "step": 12896 | |
| }, | |
| { | |
| "epoch": 124.03846153846153, | |
| "grad_norm": 7.083764553070068, | |
| "learning_rate": 9.999999999999998e-08, | |
| "loss": 10.6712, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 124.03846153846153, | |
| "eval_dummy": 1.0, | |
| "eval_loss": 33.490596771240234, | |
| "eval_runtime": 17.1907, | |
| "eval_samples_per_second": 2.792, | |
| "eval_steps_per_second": 0.582, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 124.03846153846153, | |
| "step": 12900, | |
| "total_flos": 1.8148731642810335e+19, | |
| "train_loss": 14.019830729166667, | |
| "train_runtime": 31415.0856, | |
| "train_samples_per_second": 2.053, | |
| "train_steps_per_second": 0.411 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 12900, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 125, | |
| "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": 1.8148731642810335e+19, | |
| "train_batch_size": 5, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |