| { |
| "best_metric": 3.681889772415161, |
| "best_model_checkpoint": "models/GPT2_five_function_within_boundary_53/checkpoint-64080", |
| "epoch": 10.0, |
| "global_step": 64080, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.16, |
| "learning_rate": 1e-05, |
| "loss": 7.0204, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.31, |
| "learning_rate": 2e-05, |
| "loss": 5.8208, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.47, |
| "learning_rate": 3e-05, |
| "loss": 5.4769, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.62, |
| "learning_rate": 4e-05, |
| "loss": 5.2136, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.78, |
| "learning_rate": 5e-05, |
| "loss": 5.009, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.94, |
| "learning_rate": 6e-05, |
| "loss": 4.8403, |
| "step": 6000 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.30788203681670523, |
| "eval_loss": 4.656326770782471, |
| "eval_runtime": 1.9508, |
| "eval_samples_per_second": 603.859, |
| "eval_steps_per_second": 5.126, |
| "step": 6408 |
| }, |
| { |
| "epoch": 1.09, |
| "learning_rate": 7e-05, |
| "loss": 4.6954, |
| "step": 7000 |
| }, |
| { |
| "epoch": 1.25, |
| "learning_rate": 8e-05, |
| "loss": 4.5752, |
| "step": 8000 |
| }, |
| { |
| "epoch": 1.4, |
| "learning_rate": 9e-05, |
| "loss": 4.4783, |
| "step": 9000 |
| }, |
| { |
| "epoch": 1.56, |
| "learning_rate": 0.0001, |
| "loss": 4.3997, |
| "step": 10000 |
| }, |
| { |
| "epoch": 1.72, |
| "learning_rate": 9.815273668639054e-05, |
| "loss": 4.3301, |
| "step": 11000 |
| }, |
| { |
| "epoch": 1.87, |
| "learning_rate": 9.630362426035503e-05, |
| "loss": 4.2687, |
| "step": 12000 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.3457882705239095, |
| "eval_loss": 4.142186641693115, |
| "eval_runtime": 1.9965, |
| "eval_samples_per_second": 590.033, |
| "eval_steps_per_second": 5.009, |
| "step": 12816 |
| }, |
| { |
| "epoch": 2.03, |
| "learning_rate": 9.445636094674557e-05, |
| "loss": 4.2108, |
| "step": 13000 |
| }, |
| { |
| "epoch": 2.18, |
| "learning_rate": 9.260724852071006e-05, |
| "loss": 4.1402, |
| "step": 14000 |
| }, |
| { |
| "epoch": 2.34, |
| "learning_rate": 9.07599852071006e-05, |
| "loss": 4.1108, |
| "step": 15000 |
| }, |
| { |
| "epoch": 2.5, |
| "learning_rate": 8.891087278106509e-05, |
| "loss": 4.085, |
| "step": 16000 |
| }, |
| { |
| "epoch": 2.65, |
| "learning_rate": 8.706360946745562e-05, |
| "loss": 4.0586, |
| "step": 17000 |
| }, |
| { |
| "epoch": 2.81, |
| "learning_rate": 8.521634615384617e-05, |
| "loss": 4.0369, |
| "step": 18000 |
| }, |
| { |
| "epoch": 2.97, |
| "learning_rate": 8.336723372781065e-05, |
| "loss": 4.016, |
| "step": 19000 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.3632073579936634, |
| "eval_loss": 3.9519155025482178, |
| "eval_runtime": 2.1983, |
| "eval_samples_per_second": 535.863, |
| "eval_steps_per_second": 4.549, |
| "step": 19224 |
| }, |
| { |
| "epoch": 3.12, |
| "learning_rate": 8.151997041420119e-05, |
| "loss": 3.952, |
| "step": 20000 |
| }, |
| { |
| "epoch": 3.28, |
| "learning_rate": 7.967085798816568e-05, |
| "loss": 3.9319, |
| "step": 21000 |
| }, |
| { |
| "epoch": 3.43, |
| "learning_rate": 7.782359467455622e-05, |
| "loss": 3.9251, |
| "step": 22000 |
| }, |
| { |
| "epoch": 3.59, |
| "learning_rate": 7.597448224852071e-05, |
| "loss": 3.9138, |
| "step": 23000 |
| }, |
| { |
| "epoch": 3.75, |
| "learning_rate": 7.412536982248522e-05, |
| "loss": 3.9029, |
| "step": 24000 |
| }, |
| { |
| "epoch": 3.9, |
| "learning_rate": 7.227810650887574e-05, |
| "loss": 3.8939, |
| "step": 25000 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.3735277996871783, |
| "eval_loss": 3.8524959087371826, |
| "eval_runtime": 1.965, |
| "eval_samples_per_second": 599.484, |
| "eval_steps_per_second": 5.089, |
| "step": 25632 |
| }, |
| { |
| "epoch": 4.06, |
| "learning_rate": 7.042899408284024e-05, |
| "loss": 3.8591, |
| "step": 26000 |
| }, |
| { |
| "epoch": 4.21, |
| "learning_rate": 6.858173076923077e-05, |
| "loss": 3.8163, |
| "step": 27000 |
| }, |
| { |
| "epoch": 4.37, |
| "learning_rate": 6.673261834319527e-05, |
| "loss": 3.8166, |
| "step": 28000 |
| }, |
| { |
| "epoch": 4.53, |
| "learning_rate": 6.488535502958581e-05, |
| "loss": 3.8138, |
| "step": 29000 |
| }, |
| { |
| "epoch": 4.68, |
| "learning_rate": 6.30362426035503e-05, |
| "loss": 3.8106, |
| "step": 30000 |
| }, |
| { |
| "epoch": 4.84, |
| "learning_rate": 6.118897928994083e-05, |
| "loss": 3.8033, |
| "step": 31000 |
| }, |
| { |
| "epoch": 4.99, |
| "learning_rate": 5.9339866863905326e-05, |
| "loss": 3.8002, |
| "step": 32000 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.3787481785489887, |
| "eval_loss": 3.794093608856201, |
| "eval_runtime": 1.9599, |
| "eval_samples_per_second": 601.054, |
| "eval_steps_per_second": 5.102, |
| "step": 32040 |
| }, |
| { |
| "epoch": 5.15, |
| "learning_rate": 5.7492603550295866e-05, |
| "loss": 3.7321, |
| "step": 33000 |
| }, |
| { |
| "epoch": 5.31, |
| "learning_rate": 5.564349112426036e-05, |
| "loss": 3.7353, |
| "step": 34000 |
| }, |
| { |
| "epoch": 5.46, |
| "learning_rate": 5.379622781065089e-05, |
| "loss": 3.7353, |
| "step": 35000 |
| }, |
| { |
| "epoch": 5.62, |
| "learning_rate": 5.1947115384615385e-05, |
| "loss": 3.7364, |
| "step": 36000 |
| }, |
| { |
| "epoch": 5.77, |
| "learning_rate": 5.009985207100592e-05, |
| "loss": 3.7344, |
| "step": 37000 |
| }, |
| { |
| "epoch": 5.93, |
| "learning_rate": 4.8250739644970414e-05, |
| "loss": 3.7294, |
| "step": 38000 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.3842894001577477, |
| "eval_loss": 3.753204107284546, |
| "eval_runtime": 1.9672, |
| "eval_samples_per_second": 598.835, |
| "eval_steps_per_second": 5.083, |
| "step": 38448 |
| }, |
| { |
| "epoch": 6.09, |
| "learning_rate": 4.6403476331360954e-05, |
| "loss": 3.6926, |
| "step": 39000 |
| }, |
| { |
| "epoch": 6.24, |
| "learning_rate": 4.455436390532545e-05, |
| "loss": 3.6689, |
| "step": 40000 |
| }, |
| { |
| "epoch": 6.4, |
| "learning_rate": 4.2705251479289946e-05, |
| "loss": 3.673, |
| "step": 41000 |
| }, |
| { |
| "epoch": 6.55, |
| "learning_rate": 4.085798816568048e-05, |
| "loss": 3.6735, |
| "step": 42000 |
| }, |
| { |
| "epoch": 6.71, |
| "learning_rate": 3.9008875739644975e-05, |
| "loss": 3.6737, |
| "step": 43000 |
| }, |
| { |
| "epoch": 6.87, |
| "learning_rate": 3.71616124260355e-05, |
| "loss": 3.6702, |
| "step": 44000 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.38760477520955045, |
| "eval_loss": 3.7276713848114014, |
| "eval_runtime": 1.9557, |
| "eval_samples_per_second": 602.357, |
| "eval_steps_per_second": 5.113, |
| "step": 44856 |
| }, |
| { |
| "epoch": 7.02, |
| "learning_rate": 3.5312500000000005e-05, |
| "loss": 3.662, |
| "step": 45000 |
| }, |
| { |
| "epoch": 7.18, |
| "learning_rate": 3.346523668639053e-05, |
| "loss": 3.6119, |
| "step": 46000 |
| }, |
| { |
| "epoch": 7.33, |
| "learning_rate": 3.161612426035503e-05, |
| "loss": 3.6179, |
| "step": 47000 |
| }, |
| { |
| "epoch": 7.49, |
| "learning_rate": 2.9768860946745564e-05, |
| "loss": 3.6202, |
| "step": 48000 |
| }, |
| { |
| "epoch": 7.65, |
| "learning_rate": 2.791974852071006e-05, |
| "loss": 3.6203, |
| "step": 49000 |
| }, |
| { |
| "epoch": 7.8, |
| "learning_rate": 2.6070636094674556e-05, |
| "loss": 3.6217, |
| "step": 50000 |
| }, |
| { |
| "epoch": 7.96, |
| "learning_rate": 2.422337278106509e-05, |
| "loss": 3.6188, |
| "step": 51000 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.3901380960656658, |
| "eval_loss": 3.705106735229492, |
| "eval_runtime": 1.9845, |
| "eval_samples_per_second": 593.586, |
| "eval_steps_per_second": 5.039, |
| "step": 51264 |
| }, |
| { |
| "epoch": 8.11, |
| "learning_rate": 2.2374260355029585e-05, |
| "loss": 3.5778, |
| "step": 52000 |
| }, |
| { |
| "epoch": 8.27, |
| "learning_rate": 2.052699704142012e-05, |
| "loss": 3.5713, |
| "step": 53000 |
| }, |
| { |
| "epoch": 8.43, |
| "learning_rate": 1.8677884615384617e-05, |
| "loss": 3.5728, |
| "step": 54000 |
| }, |
| { |
| "epoch": 8.58, |
| "learning_rate": 1.6830621301775147e-05, |
| "loss": 3.5739, |
| "step": 55000 |
| }, |
| { |
| "epoch": 8.74, |
| "learning_rate": 1.4981508875739645e-05, |
| "loss": 3.5734, |
| "step": 56000 |
| }, |
| { |
| "epoch": 8.9, |
| "learning_rate": 1.3132396449704143e-05, |
| "loss": 3.5761, |
| "step": 57000 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.39265804847399166, |
| "eval_loss": 3.687875509262085, |
| "eval_runtime": 1.9674, |
| "eval_samples_per_second": 598.759, |
| "eval_steps_per_second": 5.083, |
| "step": 57672 |
| }, |
| { |
| "epoch": 9.05, |
| "learning_rate": 1.1285133136094674e-05, |
| "loss": 3.5587, |
| "step": 58000 |
| }, |
| { |
| "epoch": 9.21, |
| "learning_rate": 9.436020710059172e-06, |
| "loss": 3.5346, |
| "step": 59000 |
| }, |
| { |
| "epoch": 9.36, |
| "learning_rate": 7.58690828402367e-06, |
| "loss": 3.5361, |
| "step": 60000 |
| }, |
| { |
| "epoch": 9.52, |
| "learning_rate": 5.737795857988166e-06, |
| "loss": 3.5354, |
| "step": 61000 |
| }, |
| { |
| "epoch": 9.68, |
| "learning_rate": 3.890532544378699e-06, |
| "loss": 3.5352, |
| "step": 62000 |
| }, |
| { |
| "epoch": 9.83, |
| "learning_rate": 2.0414201183431953e-06, |
| "loss": 3.5336, |
| "step": 63000 |
| }, |
| { |
| "epoch": 9.99, |
| "learning_rate": 1.9415680473372782e-07, |
| "loss": 3.5318, |
| "step": 64000 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_accuracy": 0.3937007874015748, |
| "eval_loss": 3.681889772415161, |
| "eval_runtime": 1.9701, |
| "eval_samples_per_second": 597.924, |
| "eval_steps_per_second": 5.076, |
| "step": 64080 |
| }, |
| { |
| "epoch": 10.0, |
| "step": 64080, |
| "total_flos": 5.3573097259008e+17, |
| "train_loss": 3.9848127858022626, |
| "train_runtime": 27947.1316, |
| "train_samples_per_second": 293.456, |
| "train_steps_per_second": 2.293 |
| } |
| ], |
| "max_steps": 64080, |
| "num_train_epochs": 10, |
| "total_flos": 5.3573097259008e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
