| { | |
| "best_metric": 0.8732394366197183, | |
| "best_model_checkpoint": "deit-base-distilled-patch16-224-65-fold3/checkpoint-276", | |
| "epoch": 92.3076923076923, | |
| "eval_steps": 500, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9230769230769231, | |
| "eval_accuracy": 0.5492957746478874, | |
| "eval_loss": 0.7402985692024231, | |
| "eval_runtime": 0.9976, | |
| "eval_samples_per_second": 71.17, | |
| "eval_steps_per_second": 3.007, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "eval_accuracy": 0.5211267605633803, | |
| "eval_loss": 0.7199354767799377, | |
| "eval_runtime": 0.9611, | |
| "eval_samples_per_second": 73.873, | |
| "eval_steps_per_second": 3.121, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "eval_accuracy": 0.5633802816901409, | |
| "eval_loss": 0.7111433148384094, | |
| "eval_runtime": 0.9671, | |
| "eval_samples_per_second": 73.414, | |
| "eval_steps_per_second": 3.102, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 7.598382949829102, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.7693, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5352112676056338, | |
| "eval_loss": 0.7014583945274353, | |
| "eval_runtime": 0.9614, | |
| "eval_samples_per_second": 73.854, | |
| "eval_steps_per_second": 3.121, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "eval_accuracy": 0.6197183098591549, | |
| "eval_loss": 0.6470862627029419, | |
| "eval_runtime": 0.965, | |
| "eval_samples_per_second": 73.577, | |
| "eval_steps_per_second": 3.109, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.846153846153846, | |
| "eval_accuracy": 0.6056338028169014, | |
| "eval_loss": 0.6691372394561768, | |
| "eval_runtime": 0.966, | |
| "eval_samples_per_second": 73.497, | |
| "eval_steps_per_second": 3.105, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 1.975386381149292, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.6542, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.769230769230769, | |
| "eval_accuracy": 0.6197183098591549, | |
| "eval_loss": 0.6188070774078369, | |
| "eval_runtime": 0.9657, | |
| "eval_samples_per_second": 73.52, | |
| "eval_steps_per_second": 3.106, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.5774647887323944, | |
| "eval_loss": 0.69666588306427, | |
| "eval_runtime": 0.9856, | |
| "eval_samples_per_second": 72.037, | |
| "eval_steps_per_second": 3.044, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 8.923076923076923, | |
| "eval_accuracy": 0.7323943661971831, | |
| "eval_loss": 0.5731548070907593, | |
| "eval_runtime": 0.9661, | |
| "eval_samples_per_second": 73.491, | |
| "eval_steps_per_second": 3.105, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 5.027002811431885, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5935, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 9.846153846153847, | |
| "eval_accuracy": 0.704225352112676, | |
| "eval_loss": 0.5184200406074524, | |
| "eval_runtime": 0.9721, | |
| "eval_samples_per_second": 73.035, | |
| "eval_steps_per_second": 3.086, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 10.76923076923077, | |
| "eval_accuracy": 0.7183098591549296, | |
| "eval_loss": 0.6031274795532227, | |
| "eval_runtime": 0.976, | |
| "eval_samples_per_second": 72.75, | |
| "eval_steps_per_second": 3.074, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.647887323943662, | |
| "eval_loss": 0.6670598983764648, | |
| "eval_runtime": 0.9859, | |
| "eval_samples_per_second": 72.016, | |
| "eval_steps_per_second": 3.043, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 12.307692307692308, | |
| "grad_norm": 7.941720008850098, | |
| "learning_rate": 4.814814814814815e-05, | |
| "loss": 0.549, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 12.923076923076923, | |
| "eval_accuracy": 0.7183098591549296, | |
| "eval_loss": 0.5281336903572083, | |
| "eval_runtime": 1.0091, | |
| "eval_samples_per_second": 70.358, | |
| "eval_steps_per_second": 2.973, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 13.846153846153847, | |
| "eval_accuracy": 0.7183098591549296, | |
| "eval_loss": 0.5792147517204285, | |
| "eval_runtime": 1.0732, | |
| "eval_samples_per_second": 66.158, | |
| "eval_steps_per_second": 2.795, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 14.76923076923077, | |
| "eval_accuracy": 0.7464788732394366, | |
| "eval_loss": 0.5389305353164673, | |
| "eval_runtime": 1.015, | |
| "eval_samples_per_second": 69.952, | |
| "eval_steps_per_second": 2.956, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 15.384615384615385, | |
| "grad_norm": 3.035529613494873, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.4778, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.704225352112676, | |
| "eval_loss": 0.600963830947876, | |
| "eval_runtime": 0.9862, | |
| "eval_samples_per_second": 71.992, | |
| "eval_steps_per_second": 3.042, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 16.923076923076923, | |
| "eval_accuracy": 0.7605633802816901, | |
| "eval_loss": 0.5244532227516174, | |
| "eval_runtime": 0.9743, | |
| "eval_samples_per_second": 72.876, | |
| "eval_steps_per_second": 3.079, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 17.846153846153847, | |
| "eval_accuracy": 0.7183098591549296, | |
| "eval_loss": 0.5491234064102173, | |
| "eval_runtime": 0.997, | |
| "eval_samples_per_second": 71.212, | |
| "eval_steps_per_second": 3.009, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 18.46153846153846, | |
| "grad_norm": 2.951366662979126, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.4039, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 18.76923076923077, | |
| "eval_accuracy": 0.7464788732394366, | |
| "eval_loss": 0.55897057056427, | |
| "eval_runtime": 0.9926, | |
| "eval_samples_per_second": 71.527, | |
| "eval_steps_per_second": 3.022, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7323943661971831, | |
| "eval_loss": 0.4886135458946228, | |
| "eval_runtime": 0.9968, | |
| "eval_samples_per_second": 71.225, | |
| "eval_steps_per_second": 3.01, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 20.923076923076923, | |
| "eval_accuracy": 0.7323943661971831, | |
| "eval_loss": 0.5049741268157959, | |
| "eval_runtime": 0.9885, | |
| "eval_samples_per_second": 71.822, | |
| "eval_steps_per_second": 3.035, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 21.53846153846154, | |
| "grad_norm": 2.950965404510498, | |
| "learning_rate": 4.259259259259259e-05, | |
| "loss": 0.3409, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 21.846153846153847, | |
| "eval_accuracy": 0.7464788732394366, | |
| "eval_loss": 0.49122846126556396, | |
| "eval_runtime": 0.9814, | |
| "eval_samples_per_second": 72.349, | |
| "eval_steps_per_second": 3.057, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 22.76923076923077, | |
| "eval_accuracy": 0.7746478873239436, | |
| "eval_loss": 0.492927223443985, | |
| "eval_runtime": 0.9891, | |
| "eval_samples_per_second": 71.783, | |
| "eval_steps_per_second": 3.033, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.7746478873239436, | |
| "eval_loss": 0.5365388989448547, | |
| "eval_runtime": 0.9902, | |
| "eval_samples_per_second": 71.704, | |
| "eval_steps_per_second": 3.03, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 24.615384615384617, | |
| "grad_norm": 8.019288063049316, | |
| "learning_rate": 4.074074074074074e-05, | |
| "loss": 0.3202, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 24.923076923076923, | |
| "eval_accuracy": 0.8028169014084507, | |
| "eval_loss": 0.4685072600841522, | |
| "eval_runtime": 0.983, | |
| "eval_samples_per_second": 72.224, | |
| "eval_steps_per_second": 3.052, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 25.846153846153847, | |
| "eval_accuracy": 0.8169014084507042, | |
| "eval_loss": 0.44036272168159485, | |
| "eval_runtime": 0.9931, | |
| "eval_samples_per_second": 71.493, | |
| "eval_steps_per_second": 3.021, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 26.76923076923077, | |
| "eval_accuracy": 0.8028169014084507, | |
| "eval_loss": 0.463866263628006, | |
| "eval_runtime": 0.9949, | |
| "eval_samples_per_second": 71.364, | |
| "eval_steps_per_second": 3.015, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 27.692307692307693, | |
| "grad_norm": 2.1704952716827393, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.2466, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.7605633802816901, | |
| "eval_loss": 0.5491413474082947, | |
| "eval_runtime": 0.9936, | |
| "eval_samples_per_second": 71.461, | |
| "eval_steps_per_second": 3.019, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 28.923076923076923, | |
| "eval_accuracy": 0.7605633802816901, | |
| "eval_loss": 0.5169638991355896, | |
| "eval_runtime": 1.0004, | |
| "eval_samples_per_second": 70.97, | |
| "eval_steps_per_second": 2.999, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 29.846153846153847, | |
| "eval_accuracy": 0.8028169014084507, | |
| "eval_loss": 0.4444153606891632, | |
| "eval_runtime": 0.9937, | |
| "eval_samples_per_second": 71.452, | |
| "eval_steps_per_second": 3.019, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 30.76923076923077, | |
| "grad_norm": 3.7423977851867676, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.2433, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 30.76923076923077, | |
| "eval_accuracy": 0.8309859154929577, | |
| "eval_loss": 0.4516857862472534, | |
| "eval_runtime": 0.9934, | |
| "eval_samples_per_second": 71.473, | |
| "eval_steps_per_second": 3.02, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7605633802816901, | |
| "eval_loss": 0.7796855568885803, | |
| "eval_runtime": 0.9972, | |
| "eval_samples_per_second": 71.202, | |
| "eval_steps_per_second": 3.009, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 32.92307692307692, | |
| "eval_accuracy": 0.8169014084507042, | |
| "eval_loss": 0.43210193514823914, | |
| "eval_runtime": 0.9973, | |
| "eval_samples_per_second": 71.192, | |
| "eval_steps_per_second": 3.008, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 33.84615384615385, | |
| "grad_norm": 4.042857646942139, | |
| "learning_rate": 3.518518518518519e-05, | |
| "loss": 0.2535, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 33.84615384615385, | |
| "eval_accuracy": 0.7746478873239436, | |
| "eval_loss": 0.5956353545188904, | |
| "eval_runtime": 0.9936, | |
| "eval_samples_per_second": 71.457, | |
| "eval_steps_per_second": 3.019, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 34.76923076923077, | |
| "eval_accuracy": 0.7887323943661971, | |
| "eval_loss": 0.46952158212661743, | |
| "eval_runtime": 1.0048, | |
| "eval_samples_per_second": 70.661, | |
| "eval_steps_per_second": 2.986, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.6901408450704225, | |
| "eval_loss": 0.8105931878089905, | |
| "eval_runtime": 0.9983, | |
| "eval_samples_per_second": 71.118, | |
| "eval_steps_per_second": 3.005, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 36.92307692307692, | |
| "grad_norm": 3.739529609680176, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.2215, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 36.92307692307692, | |
| "eval_accuracy": 0.7464788732394366, | |
| "eval_loss": 0.7118771076202393, | |
| "eval_runtime": 0.9979, | |
| "eval_samples_per_second": 71.147, | |
| "eval_steps_per_second": 3.006, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 37.84615384615385, | |
| "eval_accuracy": 0.8028169014084507, | |
| "eval_loss": 0.4752153158187866, | |
| "eval_runtime": 0.9912, | |
| "eval_samples_per_second": 71.631, | |
| "eval_steps_per_second": 3.027, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 38.76923076923077, | |
| "eval_accuracy": 0.8169014084507042, | |
| "eval_loss": 0.4784373342990875, | |
| "eval_runtime": 0.9957, | |
| "eval_samples_per_second": 71.31, | |
| "eval_steps_per_second": 3.013, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "grad_norm": 3.5269320011138916, | |
| "learning_rate": 3.148148148148148e-05, | |
| "loss": 0.2143, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 40.0, | |
| "eval_accuracy": 0.8028169014084507, | |
| "eval_loss": 0.4772735834121704, | |
| "eval_runtime": 1.0003, | |
| "eval_samples_per_second": 70.981, | |
| "eval_steps_per_second": 2.999, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 40.92307692307692, | |
| "eval_accuracy": 0.8169014084507042, | |
| "eval_loss": 0.5580780506134033, | |
| "eval_runtime": 0.9949, | |
| "eval_samples_per_second": 71.366, | |
| "eval_steps_per_second": 3.015, | |
| "step": 133 | |
| }, | |
| { | |
| "epoch": 41.84615384615385, | |
| "eval_accuracy": 0.8028169014084507, | |
| "eval_loss": 0.6098394989967346, | |
| "eval_runtime": 0.9964, | |
| "eval_samples_per_second": 71.254, | |
| "eval_steps_per_second": 3.011, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 42.76923076923077, | |
| "eval_accuracy": 0.8169014084507042, | |
| "eval_loss": 0.5193454623222351, | |
| "eval_runtime": 0.9881, | |
| "eval_samples_per_second": 71.858, | |
| "eval_steps_per_second": 3.036, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 43.07692307692308, | |
| "grad_norm": 2.791985273361206, | |
| "learning_rate": 2.962962962962963e-05, | |
| "loss": 0.1726, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 44.0, | |
| "eval_accuracy": 0.8450704225352113, | |
| "eval_loss": 0.4306134879589081, | |
| "eval_runtime": 0.9976, | |
| "eval_samples_per_second": 71.173, | |
| "eval_steps_per_second": 3.007, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 44.92307692307692, | |
| "eval_accuracy": 0.8591549295774648, | |
| "eval_loss": 0.4233701527118683, | |
| "eval_runtime": 0.9983, | |
| "eval_samples_per_second": 71.12, | |
| "eval_steps_per_second": 3.005, | |
| "step": 146 | |
| }, | |
| { | |
| "epoch": 45.84615384615385, | |
| "eval_accuracy": 0.8169014084507042, | |
| "eval_loss": 0.5263967514038086, | |
| "eval_runtime": 0.9942, | |
| "eval_samples_per_second": 71.411, | |
| "eval_steps_per_second": 3.017, | |
| "step": 149 | |
| }, | |
| { | |
| "epoch": 46.15384615384615, | |
| "grad_norm": 3.0404417514801025, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.1684, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 46.76923076923077, | |
| "eval_accuracy": 0.8028169014084507, | |
| "eval_loss": 0.7302807569503784, | |
| "eval_runtime": 0.9944, | |
| "eval_samples_per_second": 71.403, | |
| "eval_steps_per_second": 3.017, | |
| "step": 152 | |
| }, | |
| { | |
| "epoch": 48.0, | |
| "eval_accuracy": 0.8169014084507042, | |
| "eval_loss": 0.5079183578491211, | |
| "eval_runtime": 0.993, | |
| "eval_samples_per_second": 71.499, | |
| "eval_steps_per_second": 3.021, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 48.92307692307692, | |
| "eval_accuracy": 0.8169014084507042, | |
| "eval_loss": 0.5391997694969177, | |
| "eval_runtime": 0.9941, | |
| "eval_samples_per_second": 71.421, | |
| "eval_steps_per_second": 3.018, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 49.23076923076923, | |
| "grad_norm": 5.0057854652404785, | |
| "learning_rate": 2.5925925925925925e-05, | |
| "loss": 0.1604, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 49.84615384615385, | |
| "eval_accuracy": 0.8169014084507042, | |
| "eval_loss": 0.3951334059238434, | |
| "eval_runtime": 0.9969, | |
| "eval_samples_per_second": 71.222, | |
| "eval_steps_per_second": 3.009, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 50.76923076923077, | |
| "eval_accuracy": 0.8028169014084507, | |
| "eval_loss": 0.43108686804771423, | |
| "eval_runtime": 1.0007, | |
| "eval_samples_per_second": 70.95, | |
| "eval_steps_per_second": 2.998, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 52.0, | |
| "eval_accuracy": 0.8028169014084507, | |
| "eval_loss": 0.40819185972213745, | |
| "eval_runtime": 1.0027, | |
| "eval_samples_per_second": 70.81, | |
| "eval_steps_per_second": 2.992, | |
| "step": 169 | |
| }, | |
| { | |
| "epoch": 52.30769230769231, | |
| "grad_norm": 2.250169515609741, | |
| "learning_rate": 2.4074074074074074e-05, | |
| "loss": 0.1457, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 52.92307692307692, | |
| "eval_accuracy": 0.7887323943661971, | |
| "eval_loss": 0.4173259139060974, | |
| "eval_runtime": 1.0625, | |
| "eval_samples_per_second": 66.825, | |
| "eval_steps_per_second": 2.824, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 53.84615384615385, | |
| "eval_accuracy": 0.8309859154929577, | |
| "eval_loss": 0.4310729205608368, | |
| "eval_runtime": 1.0019, | |
| "eval_samples_per_second": 70.864, | |
| "eval_steps_per_second": 2.994, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 54.76923076923077, | |
| "eval_accuracy": 0.8028169014084507, | |
| "eval_loss": 0.42128264904022217, | |
| "eval_runtime": 1.0021, | |
| "eval_samples_per_second": 70.851, | |
| "eval_steps_per_second": 2.994, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 55.38461538461539, | |
| "grad_norm": 3.804135799407959, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.1549, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 56.0, | |
| "eval_accuracy": 0.8450704225352113, | |
| "eval_loss": 0.4712894558906555, | |
| "eval_runtime": 0.9989, | |
| "eval_samples_per_second": 71.076, | |
| "eval_steps_per_second": 3.003, | |
| "step": 182 | |
| }, | |
| { | |
| "epoch": 56.92307692307692, | |
| "eval_accuracy": 0.8028169014084507, | |
| "eval_loss": 0.749286949634552, | |
| "eval_runtime": 0.9921, | |
| "eval_samples_per_second": 71.569, | |
| "eval_steps_per_second": 3.024, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 57.84615384615385, | |
| "eval_accuracy": 0.8450704225352113, | |
| "eval_loss": 0.5160987973213196, | |
| "eval_runtime": 0.9878, | |
| "eval_samples_per_second": 71.876, | |
| "eval_steps_per_second": 3.037, | |
| "step": 188 | |
| }, | |
| { | |
| "epoch": 58.46153846153846, | |
| "grad_norm": 2.6206445693969727, | |
| "learning_rate": 2.037037037037037e-05, | |
| "loss": 0.1391, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 58.76923076923077, | |
| "eval_accuracy": 0.8169014084507042, | |
| "eval_loss": 0.46853163838386536, | |
| "eval_runtime": 0.9872, | |
| "eval_samples_per_second": 71.917, | |
| "eval_steps_per_second": 3.039, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 60.0, | |
| "eval_accuracy": 0.8028169014084507, | |
| "eval_loss": 0.6968369483947754, | |
| "eval_runtime": 0.9894, | |
| "eval_samples_per_second": 71.763, | |
| "eval_steps_per_second": 3.032, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 60.92307692307692, | |
| "eval_accuracy": 0.8309859154929577, | |
| "eval_loss": 0.5836513638496399, | |
| "eval_runtime": 0.9857, | |
| "eval_samples_per_second": 72.029, | |
| "eval_steps_per_second": 3.043, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 61.53846153846154, | |
| "grad_norm": 1.732333779335022, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1272, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 61.84615384615385, | |
| "eval_accuracy": 0.8169014084507042, | |
| "eval_loss": 0.5863271951675415, | |
| "eval_runtime": 0.9938, | |
| "eval_samples_per_second": 71.442, | |
| "eval_steps_per_second": 3.019, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 62.76923076923077, | |
| "eval_accuracy": 0.8309859154929577, | |
| "eval_loss": 0.5459548830986023, | |
| "eval_runtime": 0.9852, | |
| "eval_samples_per_second": 72.065, | |
| "eval_steps_per_second": 3.045, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 64.0, | |
| "eval_accuracy": 0.8309859154929577, | |
| "eval_loss": 0.6198180317878723, | |
| "eval_runtime": 0.9942, | |
| "eval_samples_per_second": 71.415, | |
| "eval_steps_per_second": 3.018, | |
| "step": 208 | |
| }, | |
| { | |
| "epoch": 64.61538461538461, | |
| "grad_norm": 3.675976514816284, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.1341, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 64.92307692307692, | |
| "eval_accuracy": 0.8591549295774648, | |
| "eval_loss": 0.558370053768158, | |
| "eval_runtime": 1.01, | |
| "eval_samples_per_second": 70.294, | |
| "eval_steps_per_second": 2.97, | |
| "step": 211 | |
| }, | |
| { | |
| "epoch": 65.84615384615384, | |
| "eval_accuracy": 0.8450704225352113, | |
| "eval_loss": 0.6429418325424194, | |
| "eval_runtime": 0.992, | |
| "eval_samples_per_second": 71.571, | |
| "eval_steps_per_second": 3.024, | |
| "step": 214 | |
| }, | |
| { | |
| "epoch": 66.76923076923077, | |
| "eval_accuracy": 0.8028169014084507, | |
| "eval_loss": 0.8592150211334229, | |
| "eval_runtime": 1.0035, | |
| "eval_samples_per_second": 70.751, | |
| "eval_steps_per_second": 2.989, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 67.6923076923077, | |
| "grad_norm": 1.6102505922317505, | |
| "learning_rate": 1.4814814814814815e-05, | |
| "loss": 0.1144, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 68.0, | |
| "eval_accuracy": 0.8028169014084507, | |
| "eval_loss": 0.8471563458442688, | |
| "eval_runtime": 0.989, | |
| "eval_samples_per_second": 71.787, | |
| "eval_steps_per_second": 3.033, | |
| "step": 221 | |
| }, | |
| { | |
| "epoch": 68.92307692307692, | |
| "eval_accuracy": 0.8169014084507042, | |
| "eval_loss": 0.83599853515625, | |
| "eval_runtime": 0.9959, | |
| "eval_samples_per_second": 71.291, | |
| "eval_steps_per_second": 3.012, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 69.84615384615384, | |
| "eval_accuracy": 0.8169014084507042, | |
| "eval_loss": 0.6696776151657104, | |
| "eval_runtime": 0.9955, | |
| "eval_samples_per_second": 71.323, | |
| "eval_steps_per_second": 3.014, | |
| "step": 227 | |
| }, | |
| { | |
| "epoch": 70.76923076923077, | |
| "grad_norm": 2.509876012802124, | |
| "learning_rate": 1.2962962962962962e-05, | |
| "loss": 0.1321, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 70.76923076923077, | |
| "eval_accuracy": 0.8028169014084507, | |
| "eval_loss": 0.6625356674194336, | |
| "eval_runtime": 0.9946, | |
| "eval_samples_per_second": 71.389, | |
| "eval_steps_per_second": 3.016, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 72.0, | |
| "eval_accuracy": 0.8309859154929577, | |
| "eval_loss": 0.7227580547332764, | |
| "eval_runtime": 0.9869, | |
| "eval_samples_per_second": 71.944, | |
| "eval_steps_per_second": 3.04, | |
| "step": 234 | |
| }, | |
| { | |
| "epoch": 72.92307692307692, | |
| "eval_accuracy": 0.8309859154929577, | |
| "eval_loss": 0.6792935729026794, | |
| "eval_runtime": 0.9957, | |
| "eval_samples_per_second": 71.304, | |
| "eval_steps_per_second": 3.013, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 73.84615384615384, | |
| "grad_norm": 3.0507752895355225, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.1206, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 73.84615384615384, | |
| "eval_accuracy": 0.8591549295774648, | |
| "eval_loss": 0.5570999979972839, | |
| "eval_runtime": 0.9893, | |
| "eval_samples_per_second": 71.771, | |
| "eval_steps_per_second": 3.033, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 74.76923076923077, | |
| "eval_accuracy": 0.8450704225352113, | |
| "eval_loss": 0.5105892419815063, | |
| "eval_runtime": 1.0212, | |
| "eval_samples_per_second": 69.527, | |
| "eval_steps_per_second": 2.938, | |
| "step": 243 | |
| }, | |
| { | |
| "epoch": 76.0, | |
| "eval_accuracy": 0.8309859154929577, | |
| "eval_loss": 0.6685615181922913, | |
| "eval_runtime": 1.0036, | |
| "eval_samples_per_second": 70.748, | |
| "eval_steps_per_second": 2.989, | |
| "step": 247 | |
| }, | |
| { | |
| "epoch": 76.92307692307692, | |
| "grad_norm": 3.6703972816467285, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.131, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 76.92307692307692, | |
| "eval_accuracy": 0.8309859154929577, | |
| "eval_loss": 0.7132428288459778, | |
| "eval_runtime": 0.9879, | |
| "eval_samples_per_second": 71.873, | |
| "eval_steps_per_second": 3.037, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 77.84615384615384, | |
| "eval_accuracy": 0.8450704225352113, | |
| "eval_loss": 0.5945414900779724, | |
| "eval_runtime": 0.9943, | |
| "eval_samples_per_second": 71.404, | |
| "eval_steps_per_second": 3.017, | |
| "step": 253 | |
| }, | |
| { | |
| "epoch": 78.76923076923077, | |
| "eval_accuracy": 0.7746478873239436, | |
| "eval_loss": 0.5515730977058411, | |
| "eval_runtime": 1.0089, | |
| "eval_samples_per_second": 70.372, | |
| "eval_steps_per_second": 2.973, | |
| "step": 256 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "grad_norm": 1.8021163940429688, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 0.1009, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 80.0, | |
| "eval_accuracy": 0.7605633802816901, | |
| "eval_loss": 0.547419011592865, | |
| "eval_runtime": 1.0062, | |
| "eval_samples_per_second": 70.563, | |
| "eval_steps_per_second": 2.982, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 80.92307692307692, | |
| "eval_accuracy": 0.7887323943661971, | |
| "eval_loss": 0.5218686461448669, | |
| "eval_runtime": 0.9971, | |
| "eval_samples_per_second": 71.207, | |
| "eval_steps_per_second": 3.009, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 81.84615384615384, | |
| "eval_accuracy": 0.8450704225352113, | |
| "eval_loss": 0.5374658107757568, | |
| "eval_runtime": 0.9899, | |
| "eval_samples_per_second": 71.721, | |
| "eval_steps_per_second": 3.03, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 82.76923076923077, | |
| "eval_accuracy": 0.8450704225352113, | |
| "eval_loss": 0.5133435726165771, | |
| "eval_runtime": 0.9965, | |
| "eval_samples_per_second": 71.253, | |
| "eval_steps_per_second": 3.011, | |
| "step": 269 | |
| }, | |
| { | |
| "epoch": 83.07692307692308, | |
| "grad_norm": 1.6209955215454102, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.1084, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 84.0, | |
| "eval_accuracy": 0.8450704225352113, | |
| "eval_loss": 0.49109983444213867, | |
| "eval_runtime": 0.9882, | |
| "eval_samples_per_second": 71.848, | |
| "eval_steps_per_second": 3.036, | |
| "step": 273 | |
| }, | |
| { | |
| "epoch": 84.92307692307692, | |
| "eval_accuracy": 0.8732394366197183, | |
| "eval_loss": 0.49927499890327454, | |
| "eval_runtime": 1.0035, | |
| "eval_samples_per_second": 70.749, | |
| "eval_steps_per_second": 2.989, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 85.84615384615384, | |
| "eval_accuracy": 0.8591549295774648, | |
| "eval_loss": 0.5418108701705933, | |
| "eval_runtime": 0.9918, | |
| "eval_samples_per_second": 71.588, | |
| "eval_steps_per_second": 3.025, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 86.15384615384616, | |
| "grad_norm": 2.604188919067383, | |
| "learning_rate": 3.7037037037037037e-06, | |
| "loss": 0.0851, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 86.76923076923077, | |
| "eval_accuracy": 0.8450704225352113, | |
| "eval_loss": 0.6010194420814514, | |
| "eval_runtime": 0.998, | |
| "eval_samples_per_second": 71.143, | |
| "eval_steps_per_second": 3.006, | |
| "step": 282 | |
| }, | |
| { | |
| "epoch": 88.0, | |
| "eval_accuracy": 0.8450704225352113, | |
| "eval_loss": 0.6304585933685303, | |
| "eval_runtime": 0.9947, | |
| "eval_samples_per_second": 71.377, | |
| "eval_steps_per_second": 3.016, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 88.92307692307692, | |
| "eval_accuracy": 0.8450704225352113, | |
| "eval_loss": 0.6016303300857544, | |
| "eval_runtime": 0.9919, | |
| "eval_samples_per_second": 71.58, | |
| "eval_steps_per_second": 3.025, | |
| "step": 289 | |
| }, | |
| { | |
| "epoch": 89.23076923076923, | |
| "grad_norm": 2.3317203521728516, | |
| "learning_rate": 1.8518518518518519e-06, | |
| "loss": 0.1071, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 89.84615384615384, | |
| "eval_accuracy": 0.8591549295774648, | |
| "eval_loss": 0.577277660369873, | |
| "eval_runtime": 0.9927, | |
| "eval_samples_per_second": 71.522, | |
| "eval_steps_per_second": 3.022, | |
| "step": 292 | |
| }, | |
| { | |
| "epoch": 90.76923076923077, | |
| "eval_accuracy": 0.8732394366197183, | |
| "eval_loss": 0.5610097646713257, | |
| "eval_runtime": 1.0072, | |
| "eval_samples_per_second": 70.491, | |
| "eval_steps_per_second": 2.978, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 92.0, | |
| "eval_accuracy": 0.8732394366197183, | |
| "eval_loss": 0.5522463917732239, | |
| "eval_runtime": 0.9964, | |
| "eval_samples_per_second": 71.258, | |
| "eval_steps_per_second": 3.011, | |
| "step": 299 | |
| }, | |
| { | |
| "epoch": 92.3076923076923, | |
| "grad_norm": 2.7261242866516113, | |
| "learning_rate": 0.0, | |
| "loss": 0.1139, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 92.3076923076923, | |
| "eval_accuracy": 0.8732394366197183, | |
| "eval_loss": 0.5514280796051025, | |
| "eval_runtime": 1.0006, | |
| "eval_samples_per_second": 70.955, | |
| "eval_steps_per_second": 2.998, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 92.3076923076923, | |
| "step": 300, | |
| "total_flos": 2.8402872494292173e+18, | |
| "train_loss": 0.2501216806968053, | |
| "train_runtime": 1631.2353, | |
| "train_samples_per_second": 24.337, | |
| "train_steps_per_second": 0.184 | |
| }, | |
| { | |
| "epoch": 92.3076923076923, | |
| "eval_accuracy": 0.8732394366197183, | |
| "eval_loss": 0.49927499890327454, | |
| "eval_runtime": 1.0275, | |
| "eval_samples_per_second": 69.097, | |
| "eval_steps_per_second": 2.92, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 100, | |
| "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": 2.8402872494292173e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |