| { | |
| "best_metric": 1.0332472324371338, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-650", | |
| "epoch": 5.0, | |
| "eval_steps": 500, | |
| "global_step": 650, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07692307692307693, | |
| "grad_norm": 2.0955963134765625, | |
| "learning_rate": 1.9692307692307696e-05, | |
| "loss": 1.1239, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 1.8236017227172852, | |
| "learning_rate": 1.9384615384615386e-05, | |
| "loss": 1.1222, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23076923076923078, | |
| "grad_norm": 1.9866633415222168, | |
| "learning_rate": 1.907692307692308e-05, | |
| "loss": 1.1165, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 2.4556338787078857, | |
| "learning_rate": 1.876923076923077e-05, | |
| "loss": 1.1047, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 1.6562598943710327, | |
| "learning_rate": 1.8461538461538465e-05, | |
| "loss": 1.1085, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 1.501336932182312, | |
| "learning_rate": 1.8153846153846155e-05, | |
| "loss": 1.1048, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5384615384615384, | |
| "grad_norm": 2.606107234954834, | |
| "learning_rate": 1.784615384615385e-05, | |
| "loss": 1.1031, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 2.634584426879883, | |
| "learning_rate": 1.753846153846154e-05, | |
| "loss": 1.0916, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6923076923076923, | |
| "grad_norm": 2.039886236190796, | |
| "learning_rate": 1.7230769230769234e-05, | |
| "loss": 1.095, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.4800949096679688, | |
| "learning_rate": 1.6923076923076924e-05, | |
| "loss": 1.0916, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8461538461538461, | |
| "grad_norm": 1.6766040325164795, | |
| "learning_rate": 1.6615384615384618e-05, | |
| "loss": 1.0959, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 2.2848353385925293, | |
| "learning_rate": 1.630769230769231e-05, | |
| "loss": 1.1005, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 4.07199239730835, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.0881, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.41353383458646614, | |
| "eval_loss": 1.0901767015457153, | |
| "eval_runtime": 0.7894, | |
| "eval_samples_per_second": 168.48, | |
| "eval_steps_per_second": 21.535, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "grad_norm": 2.0711560249328613, | |
| "learning_rate": 1.5692307692307693e-05, | |
| "loss": 1.0794, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 2.499732494354248, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 1.0755, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 1.4569323062896729, | |
| "learning_rate": 1.5076923076923078e-05, | |
| "loss": 1.0807, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3076923076923077, | |
| "grad_norm": 2.351478338241577, | |
| "learning_rate": 1.4769230769230772e-05, | |
| "loss": 1.0965, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "grad_norm": 2.1514322757720947, | |
| "learning_rate": 1.4461538461538462e-05, | |
| "loss": 1.0839, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.4615384615384617, | |
| "grad_norm": 2.151601791381836, | |
| "learning_rate": 1.4153846153846156e-05, | |
| "loss": 1.0837, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 1.797500729560852, | |
| "learning_rate": 1.3846153846153847e-05, | |
| "loss": 1.0807, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6153846153846154, | |
| "grad_norm": 1.7583892345428467, | |
| "learning_rate": 1.353846153846154e-05, | |
| "loss": 1.073, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "grad_norm": 3.5006496906280518, | |
| "learning_rate": 1.3230769230769231e-05, | |
| "loss": 1.0741, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.7692307692307692, | |
| "grad_norm": 2.0179672241210938, | |
| "learning_rate": 1.2923076923076925e-05, | |
| "loss": 1.077, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 2.057086944580078, | |
| "learning_rate": 1.2615384615384616e-05, | |
| "loss": 1.0925, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 1.9710103273391724, | |
| "learning_rate": 1.230769230769231e-05, | |
| "loss": 1.0757, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.9315028190612793, | |
| "learning_rate": 1.2e-05, | |
| "loss": 1.0716, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5037593984962406, | |
| "eval_loss": 1.0685255527496338, | |
| "eval_runtime": 0.7711, | |
| "eval_samples_per_second": 172.471, | |
| "eval_steps_per_second": 22.045, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.076923076923077, | |
| "grad_norm": 2.183527946472168, | |
| "learning_rate": 1.1692307692307694e-05, | |
| "loss": 1.0786, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 2.379652500152588, | |
| "learning_rate": 1.1384615384615385e-05, | |
| "loss": 1.0676, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.230769230769231, | |
| "grad_norm": 2.387296438217163, | |
| "learning_rate": 1.1076923076923079e-05, | |
| "loss": 1.064, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 2.8164947032928467, | |
| "learning_rate": 1.076923076923077e-05, | |
| "loss": 1.0548, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.3846153846153846, | |
| "grad_norm": 1.849363088607788, | |
| "learning_rate": 1.0461538461538463e-05, | |
| "loss": 1.0693, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 2.0274300575256348, | |
| "learning_rate": 1.0153846153846154e-05, | |
| "loss": 1.0604, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.5384615384615383, | |
| "grad_norm": 1.4544223546981812, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 1.0458, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": 2.1712183952331543, | |
| "learning_rate": 9.53846153846154e-06, | |
| "loss": 1.0609, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 1.922677993774414, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 1.0646, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 1.8288860321044922, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 1.0597, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.8461538461538463, | |
| "grad_norm": 2.123480796813965, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 1.052, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "grad_norm": 1.820168137550354, | |
| "learning_rate": 8.307692307692309e-06, | |
| "loss": 1.0434, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 4.973505020141602, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.061, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6240601503759399, | |
| "eval_loss": 1.0459144115447998, | |
| "eval_runtime": 0.7618, | |
| "eval_samples_per_second": 174.588, | |
| "eval_steps_per_second": 22.316, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 2.1673223972320557, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 1.0616, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.1538461538461537, | |
| "grad_norm": 1.8567888736724854, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 1.0501, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.230769230769231, | |
| "grad_norm": 2.0640571117401123, | |
| "learning_rate": 7.076923076923078e-06, | |
| "loss": 1.0611, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.3076923076923075, | |
| "grad_norm": 2.215384006500244, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 1.0469, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "grad_norm": 2.1000049114227295, | |
| "learning_rate": 6.461538461538463e-06, | |
| "loss": 1.0499, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 1.7218382358551025, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 1.0564, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.5384615384615383, | |
| "grad_norm": 2.3569300174713135, | |
| "learning_rate": 5.846153846153847e-06, | |
| "loss": 1.0599, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.6153846153846154, | |
| "grad_norm": 1.5210909843444824, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 1.0367, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 2.7621657848358154, | |
| "learning_rate": 5.230769230769232e-06, | |
| "loss": 1.0421, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.769230769230769, | |
| "grad_norm": 1.5097808837890625, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 1.0362, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 1.5118447542190552, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 1.0572, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.9230769230769234, | |
| "grad_norm": 1.7513490915298462, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 1.0361, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 5.398025035858154, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.0514, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6015037593984962, | |
| "eval_loss": 1.0407124757766724, | |
| "eval_runtime": 0.7726, | |
| "eval_samples_per_second": 172.155, | |
| "eval_steps_per_second": 22.005, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.076923076923077, | |
| "grad_norm": 2.5516345500946045, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 1.0529, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.153846153846154, | |
| "grad_norm": 1.6976008415222168, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 1.0472, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.230769230769231, | |
| "grad_norm": 2.5672519207000732, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 1.0565, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "grad_norm": 2.166529655456543, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 1.0619, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.384615384615385, | |
| "grad_norm": 1.961472511291504, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 1.0322, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.461538461538462, | |
| "grad_norm": 2.392319440841675, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 1.0388, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.538461538461538, | |
| "grad_norm": 2.3034205436706543, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 1.0358, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 2.037050247192383, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 1.0334, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.6923076923076925, | |
| "grad_norm": 3.0737335681915283, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 1.0506, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.769230769230769, | |
| "grad_norm": 2.1824796199798584, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 1.0516, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.846153846153846, | |
| "grad_norm": 1.9239214658737183, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 1.0399, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 2.267302989959717, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 1.0374, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 4.4957404136657715, | |
| "learning_rate": 0.0, | |
| "loss": 1.05, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6766917293233082, | |
| "eval_loss": 1.0332472324371338, | |
| "eval_runtime": 0.8208, | |
| "eval_samples_per_second": 162.035, | |
| "eval_steps_per_second": 20.711, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 650, | |
| "total_flos": 5.219455168198656e+16, | |
| "train_loss": 1.0677934587918796, | |
| "train_runtime": 48.1154, | |
| "train_samples_per_second": 107.45, | |
| "train_steps_per_second": 13.509 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 650, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "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": 5.219455168198656e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |