| { | |
| "best_metric": 0.8347006440162659, | |
| "best_model_checkpoint": "./beans_lora_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": 1.496038794517517, | |
| "learning_rate": 1.9692307692307696e-05, | |
| "loss": 1.0741, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 1.14773690700531, | |
| "learning_rate": 1.9384615384615386e-05, | |
| "loss": 1.0878, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23076923076923078, | |
| "grad_norm": 1.61867356300354, | |
| "learning_rate": 1.907692307692308e-05, | |
| "loss": 1.08, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 2.2048051357269287, | |
| "learning_rate": 1.876923076923077e-05, | |
| "loss": 1.049, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 1.3817859888076782, | |
| "learning_rate": 1.8461538461538465e-05, | |
| "loss": 1.0672, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 1.385685682296753, | |
| "learning_rate": 1.8153846153846155e-05, | |
| "loss": 1.0604, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5384615384615384, | |
| "grad_norm": 1.9848027229309082, | |
| "learning_rate": 1.784615384615385e-05, | |
| "loss": 1.0445, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 1.7842564582824707, | |
| "learning_rate": 1.753846153846154e-05, | |
| "loss": 1.0431, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6923076923076923, | |
| "grad_norm": 1.729790449142456, | |
| "learning_rate": 1.7230769230769234e-05, | |
| "loss": 1.0511, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.3279037475585938, | |
| "learning_rate": 1.6923076923076924e-05, | |
| "loss": 1.0556, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8461538461538461, | |
| "grad_norm": 1.5423173904418945, | |
| "learning_rate": 1.6615384615384618e-05, | |
| "loss": 1.0575, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 1.8890392780303955, | |
| "learning_rate": 1.630769230769231e-05, | |
| "loss": 1.0304, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.8158857822418213, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.0299, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6240601503759399, | |
| "eval_loss": 1.0291587114334106, | |
| "eval_runtime": 5.0639, | |
| "eval_samples_per_second": 26.264, | |
| "eval_steps_per_second": 3.357, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "grad_norm": 1.5042554140090942, | |
| "learning_rate": 1.5692307692307693e-05, | |
| "loss": 1.0233, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 1.574145793914795, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 1.0102, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 1.3951938152313232, | |
| "learning_rate": 1.5076923076923078e-05, | |
| "loss": 1.0171, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3076923076923077, | |
| "grad_norm": 1.6762858629226685, | |
| "learning_rate": 1.4769230769230772e-05, | |
| "loss": 1.0067, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "grad_norm": 1.5464376211166382, | |
| "learning_rate": 1.4461538461538462e-05, | |
| "loss": 0.9966, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.4615384615384617, | |
| "grad_norm": 1.553270697593689, | |
| "learning_rate": 1.4153846153846156e-05, | |
| "loss": 0.998, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 1.3178517818450928, | |
| "learning_rate": 1.3846153846153847e-05, | |
| "loss": 1.0004, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6153846153846154, | |
| "grad_norm": 1.414166808128357, | |
| "learning_rate": 1.353846153846154e-05, | |
| "loss": 0.9889, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "grad_norm": 2.7710893154144287, | |
| "learning_rate": 1.3230769230769231e-05, | |
| "loss": 0.9705, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.7692307692307692, | |
| "grad_norm": 1.6936691999435425, | |
| "learning_rate": 1.2923076923076925e-05, | |
| "loss": 0.9814, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 1.4454400539398193, | |
| "learning_rate": 1.2615384615384616e-05, | |
| "loss": 0.9935, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 1.6399688720703125, | |
| "learning_rate": 1.230769230769231e-05, | |
| "loss": 0.965, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.0613536834716797, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.9616, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7443609022556391, | |
| "eval_loss": 0.9589495062828064, | |
| "eval_runtime": 4.3534, | |
| "eval_samples_per_second": 30.551, | |
| "eval_steps_per_second": 3.905, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.076923076923077, | |
| "grad_norm": 1.2324800491333008, | |
| "learning_rate": 1.1692307692307694e-05, | |
| "loss": 0.9474, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 1.842686414718628, | |
| "learning_rate": 1.1384615384615385e-05, | |
| "loss": 0.9555, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.230769230769231, | |
| "grad_norm": 1.8932175636291504, | |
| "learning_rate": 1.1076923076923079e-05, | |
| "loss": 0.9411, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 1.4860458374023438, | |
| "learning_rate": 1.076923076923077e-05, | |
| "loss": 0.931, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.3846153846153846, | |
| "grad_norm": 1.520959496498108, | |
| "learning_rate": 1.0461538461538463e-05, | |
| "loss": 0.9442, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 1.3432276248931885, | |
| "learning_rate": 1.0153846153846154e-05, | |
| "loss": 0.9292, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.5384615384615383, | |
| "grad_norm": 1.2196266651153564, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.9271, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": 1.521013855934143, | |
| "learning_rate": 9.53846153846154e-06, | |
| "loss": 0.917, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 1.5005229711532593, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 0.9225, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 1.3230764865875244, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 0.9115, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.8461538461538463, | |
| "grad_norm": 1.4843106269836426, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.9077, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "grad_norm": 1.432364583015442, | |
| "learning_rate": 8.307692307692309e-06, | |
| "loss": 0.901, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.669428586959839, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.8783, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7894736842105263, | |
| "eval_loss": 0.8891889452934265, | |
| "eval_runtime": 17.6473, | |
| "eval_samples_per_second": 7.537, | |
| "eval_steps_per_second": 0.963, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 1.462823748588562, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.9124, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.1538461538461537, | |
| "grad_norm": 1.2650939226150513, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.9026, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.230769230769231, | |
| "grad_norm": 1.514074683189392, | |
| "learning_rate": 7.076923076923078e-06, | |
| "loss": 0.9024, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.3076923076923075, | |
| "grad_norm": 2.043466091156006, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.8674, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "grad_norm": 1.8109254837036133, | |
| "learning_rate": 6.461538461538463e-06, | |
| "loss": 0.8879, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 1.5040030479431152, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.8623, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.5384615384615383, | |
| "grad_norm": 1.7636866569519043, | |
| "learning_rate": 5.846153846153847e-06, | |
| "loss": 0.8787, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.6153846153846154, | |
| "grad_norm": 1.368589162826538, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.8492, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 2.1044576168060303, | |
| "learning_rate": 5.230769230769232e-06, | |
| "loss": 0.8737, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.769230769230769, | |
| "grad_norm": 1.4423880577087402, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.8443, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 1.2733564376831055, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.8557, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.9230769230769234, | |
| "grad_norm": 1.2562549114227295, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.8574, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.079272747039795, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.836, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8045112781954887, | |
| "eval_loss": 0.8480122685432434, | |
| "eval_runtime": 5.8399, | |
| "eval_samples_per_second": 22.774, | |
| "eval_steps_per_second": 2.911, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.076923076923077, | |
| "grad_norm": 2.445186138153076, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.8466, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.153846153846154, | |
| "grad_norm": 1.5983587503433228, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 0.8842, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.230769230769231, | |
| "grad_norm": 2.1697800159454346, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.8777, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "grad_norm": 1.2998322248458862, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 0.8483, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.384615384615385, | |
| "grad_norm": 1.4422998428344727, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.8451, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.461538461538462, | |
| "grad_norm": 1.6491024494171143, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 0.8165, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.538461538461538, | |
| "grad_norm": 1.5623419284820557, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.8344, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 1.266867756843567, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.8581, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.6923076923076925, | |
| "grad_norm": 2.4471514225006104, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.8628, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.769230769230769, | |
| "grad_norm": 1.3094156980514526, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 0.8131, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.846153846153846, | |
| "grad_norm": 1.4745099544525146, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.8315, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 2.0481581687927246, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 0.8188, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.112938404083252, | |
| "learning_rate": 0.0, | |
| "loss": 0.8494, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8045112781954887, | |
| "eval_loss": 0.8347006440162659, | |
| "eval_runtime": 8.0546, | |
| "eval_samples_per_second": 16.512, | |
| "eval_steps_per_second": 2.111, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 650, | |
| "total_flos": 4.0477911662555136e+17, | |
| "train_loss": 0.9380592258159931, | |
| "train_runtime": 422.3657, | |
| "train_samples_per_second": 12.241, | |
| "train_steps_per_second": 1.539 | |
| } | |
| ], | |
| "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": 4.0477911662555136e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |