| { | |
| "best_metric": 0.8529870510101318, | |
| "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": 1.495912790298462, | |
| "learning_rate": 1.9692307692307696e-05, | |
| "loss": 1.0741, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 1.1474604606628418, | |
| "learning_rate": 1.9384615384615386e-05, | |
| "loss": 1.0879, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.23076923076923078, | |
| "grad_norm": 1.6189851760864258, | |
| "learning_rate": 1.907692307692308e-05, | |
| "loss": 1.0802, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 2.2019317150115967, | |
| "learning_rate": 1.876923076923077e-05, | |
| "loss": 1.0493, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.38461538461538464, | |
| "grad_norm": 1.3802038431167603, | |
| "learning_rate": 1.8461538461538465e-05, | |
| "loss": 1.0676, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.46153846153846156, | |
| "grad_norm": 1.385358214378357, | |
| "learning_rate": 1.8153846153846155e-05, | |
| "loss": 1.0609, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5384615384615384, | |
| "grad_norm": 1.984493374824524, | |
| "learning_rate": 1.784615384615385e-05, | |
| "loss": 1.0451, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 1.7820314168930054, | |
| "learning_rate": 1.753846153846154e-05, | |
| "loss": 1.044, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6923076923076923, | |
| "grad_norm": 1.7298434972763062, | |
| "learning_rate": 1.7230769230769234e-05, | |
| "loss": 1.0523, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 2.324188470840454, | |
| "learning_rate": 1.6923076923076924e-05, | |
| "loss": 1.0571, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.8461538461538461, | |
| "grad_norm": 1.5387742519378662, | |
| "learning_rate": 1.6615384615384618e-05, | |
| "loss": 1.0593, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 1.8900723457336426, | |
| "learning_rate": 1.630769230769231e-05, | |
| "loss": 1.0329, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 2.8200511932373047, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.0326, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6090225563909775, | |
| "eval_loss": 1.0319362878799438, | |
| "eval_runtime": 6.8087, | |
| "eval_samples_per_second": 19.534, | |
| "eval_steps_per_second": 2.497, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.0769230769230769, | |
| "grad_norm": 1.4958809614181519, | |
| "learning_rate": 1.5692307692307693e-05, | |
| "loss": 1.0263, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.1538461538461537, | |
| "grad_norm": 1.5726135969161987, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 1.0135, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.2307692307692308, | |
| "grad_norm": 1.3866033554077148, | |
| "learning_rate": 1.5076923076923078e-05, | |
| "loss": 1.0212, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.3076923076923077, | |
| "grad_norm": 1.6613730192184448, | |
| "learning_rate": 1.4769230769230772e-05, | |
| "loss": 1.0116, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.3846153846153846, | |
| "grad_norm": 1.5437824726104736, | |
| "learning_rate": 1.4461538461538462e-05, | |
| "loss": 1.0009, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.4615384615384617, | |
| "grad_norm": 1.5443662405014038, | |
| "learning_rate": 1.4153846153846156e-05, | |
| "loss": 1.0036, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 1.3017466068267822, | |
| "learning_rate": 1.3846153846153847e-05, | |
| "loss": 1.0066, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.6153846153846154, | |
| "grad_norm": 1.4007813930511475, | |
| "learning_rate": 1.353846153846154e-05, | |
| "loss": 0.9954, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.6923076923076923, | |
| "grad_norm": 2.7631592750549316, | |
| "learning_rate": 1.3230769230769231e-05, | |
| "loss": 0.9775, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.7692307692307692, | |
| "grad_norm": 1.6825937032699585, | |
| "learning_rate": 1.2923076923076925e-05, | |
| "loss": 0.9898, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "grad_norm": 1.4360311031341553, | |
| "learning_rate": 1.2615384615384616e-05, | |
| "loss": 1.0017, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.9230769230769231, | |
| "grad_norm": 1.6261993646621704, | |
| "learning_rate": 1.230769230769231e-05, | |
| "loss": 0.9741, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.090559720993042, | |
| "learning_rate": 1.2e-05, | |
| "loss": 0.9721, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.7293233082706767, | |
| "eval_loss": 0.9698564410209656, | |
| "eval_runtime": 11.0803, | |
| "eval_samples_per_second": 12.003, | |
| "eval_steps_per_second": 1.534, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.076923076923077, | |
| "grad_norm": 1.219846487045288, | |
| "learning_rate": 1.1692307692307694e-05, | |
| "loss": 0.9583, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.1538461538461537, | |
| "grad_norm": 1.85811448097229, | |
| "learning_rate": 1.1384615384615385e-05, | |
| "loss": 0.9665, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.230769230769231, | |
| "grad_norm": 1.8803926706314087, | |
| "learning_rate": 1.1076923076923079e-05, | |
| "loss": 0.9529, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 1.4712238311767578, | |
| "learning_rate": 1.076923076923077e-05, | |
| "loss": 0.9423, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.3846153846153846, | |
| "grad_norm": 1.5016744136810303, | |
| "learning_rate": 1.0461538461538463e-05, | |
| "loss": 0.9554, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.4615384615384617, | |
| "grad_norm": 1.317639946937561, | |
| "learning_rate": 1.0153846153846154e-05, | |
| "loss": 0.9414, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.5384615384615383, | |
| "grad_norm": 1.2138745784759521, | |
| "learning_rate": 9.846153846153848e-06, | |
| "loss": 0.9417, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.6153846153846154, | |
| "grad_norm": 1.5316927433013916, | |
| "learning_rate": 9.53846153846154e-06, | |
| "loss": 0.9303, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 2.6923076923076925, | |
| "grad_norm": 1.4810699224472046, | |
| "learning_rate": 9.230769230769232e-06, | |
| "loss": 0.9361, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "grad_norm": 1.3046553134918213, | |
| "learning_rate": 8.923076923076925e-06, | |
| "loss": 0.9262, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 2.8461538461538463, | |
| "grad_norm": 1.4417724609375, | |
| "learning_rate": 8.615384615384617e-06, | |
| "loss": 0.9237, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 2.9230769230769234, | |
| "grad_norm": 1.4357845783233643, | |
| "learning_rate": 8.307692307692309e-06, | |
| "loss": 0.9172, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.6237688064575195, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.8948, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.7819548872180451, | |
| "eval_loss": 0.905978798866272, | |
| "eval_runtime": 11.2154, | |
| "eval_samples_per_second": 11.859, | |
| "eval_steps_per_second": 1.516, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 1.473265290260315, | |
| "learning_rate": 7.692307692307694e-06, | |
| "loss": 0.9283, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.1538461538461537, | |
| "grad_norm": 1.2494803667068481, | |
| "learning_rate": 7.384615384615386e-06, | |
| "loss": 0.9173, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 3.230769230769231, | |
| "grad_norm": 1.5195587873458862, | |
| "learning_rate": 7.076923076923078e-06, | |
| "loss": 0.916, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 3.3076923076923075, | |
| "grad_norm": 2.023468255996704, | |
| "learning_rate": 6.76923076923077e-06, | |
| "loss": 0.8841, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 3.3846153846153846, | |
| "grad_norm": 1.7898038625717163, | |
| "learning_rate": 6.461538461538463e-06, | |
| "loss": 0.9006, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 3.4615384615384617, | |
| "grad_norm": 1.475136160850525, | |
| "learning_rate": 6.153846153846155e-06, | |
| "loss": 0.8768, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 3.5384615384615383, | |
| "grad_norm": 1.7361855506896973, | |
| "learning_rate": 5.846153846153847e-06, | |
| "loss": 0.8955, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 3.6153846153846154, | |
| "grad_norm": 1.3542256355285645, | |
| "learning_rate": 5.538461538461539e-06, | |
| "loss": 0.8628, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 3.6923076923076925, | |
| "grad_norm": 2.064852714538574, | |
| "learning_rate": 5.230769230769232e-06, | |
| "loss": 0.8895, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 3.769230769230769, | |
| "grad_norm": 1.4197514057159424, | |
| "learning_rate": 4.923076923076924e-06, | |
| "loss": 0.8612, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 1.2602348327636719, | |
| "learning_rate": 4.615384615384616e-06, | |
| "loss": 0.8698, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 3.9230769230769234, | |
| "grad_norm": 1.2403748035430908, | |
| "learning_rate": 4.307692307692308e-06, | |
| "loss": 0.8712, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 3.1204898357391357, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 0.8524, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8045112781954887, | |
| "eval_loss": 0.8660793304443359, | |
| "eval_runtime": 1.2917, | |
| "eval_samples_per_second": 102.962, | |
| "eval_steps_per_second": 13.161, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.076923076923077, | |
| "grad_norm": 2.3789713382720947, | |
| "learning_rate": 3.692307692307693e-06, | |
| "loss": 0.8624, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 4.153846153846154, | |
| "grad_norm": 1.572312831878662, | |
| "learning_rate": 3.384615384615385e-06, | |
| "loss": 0.8991, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 4.230769230769231, | |
| "grad_norm": 2.145324945449829, | |
| "learning_rate": 3.0769230769230774e-06, | |
| "loss": 0.8963, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 4.3076923076923075, | |
| "grad_norm": 1.2909295558929443, | |
| "learning_rate": 2.7692307692307697e-06, | |
| "loss": 0.8629, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 4.384615384615385, | |
| "grad_norm": 1.4236739873886108, | |
| "learning_rate": 2.461538461538462e-06, | |
| "loss": 0.8597, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 4.461538461538462, | |
| "grad_norm": 1.638335943222046, | |
| "learning_rate": 2.153846153846154e-06, | |
| "loss": 0.833, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 4.538461538461538, | |
| "grad_norm": 1.5316523313522339, | |
| "learning_rate": 1.8461538461538465e-06, | |
| "loss": 0.8521, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 1.2549222707748413, | |
| "learning_rate": 1.5384615384615387e-06, | |
| "loss": 0.8741, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 4.6923076923076925, | |
| "grad_norm": 2.406038761138916, | |
| "learning_rate": 1.230769230769231e-06, | |
| "loss": 0.8801, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 4.769230769230769, | |
| "grad_norm": 1.3158031702041626, | |
| "learning_rate": 9.230769230769232e-07, | |
| "loss": 0.829, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 4.846153846153846, | |
| "grad_norm": 1.4627388715744019, | |
| "learning_rate": 6.153846153846155e-07, | |
| "loss": 0.8492, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "grad_norm": 2.0194759368896484, | |
| "learning_rate": 3.0769230769230774e-07, | |
| "loss": 0.8351, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.150322914123535, | |
| "learning_rate": 0.0, | |
| "loss": 0.866, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8045112781954887, | |
| "eval_loss": 0.8529870510101318, | |
| "eval_runtime": 5.7804, | |
| "eval_samples_per_second": 23.009, | |
| "eval_steps_per_second": 2.941, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 650, | |
| "total_flos": 4.0340206085566464e+17, | |
| "train_loss": 0.9484451132554275, | |
| "train_runtime": 507.077, | |
| "train_samples_per_second": 10.196, | |
| "train_steps_per_second": 1.282 | |
| } | |
| ], | |
| "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": 4.0340206085566464e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |