| { | |
| "best_metric": 2.406733751296997, | |
| "best_model_checkpoint": "./gpt2_dolly_lite/checkpoint-3900", | |
| "epoch": 2.9994231878484907, | |
| "eval_steps": 500, | |
| "global_step": 3900, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 0.0009871794871794872, | |
| "loss": 3.3078, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 0.0009743589743589744, | |
| "loss": 3.1249, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 0.0009615384615384616, | |
| "loss": 3.1026, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.0009487179487179487, | |
| "loss": 3.0496, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 0.000935897435897436, | |
| "loss": 2.9706, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 0.0009230769230769232, | |
| "loss": 3.0195, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 0.0009102564102564102, | |
| "loss": 2.9747, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.0008974358974358974, | |
| "loss": 2.9205, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0008846153846153846, | |
| "loss": 2.9495, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 0.0008717948717948718, | |
| "loss": 2.8866, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 0.0008589743589743589, | |
| "loss": 2.892, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.0008461538461538462, | |
| "loss": 2.8849, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 0.0008333333333333334, | |
| "loss": 2.8819, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 0.0008205128205128205, | |
| "loss": 2.8561, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "learning_rate": 0.0008076923076923078, | |
| "loss": 2.8311, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.0007948717948717948, | |
| "loss": 2.8176, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 0.000782051282051282, | |
| "loss": 2.7782, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 0.0007692307692307693, | |
| "loss": 2.7273, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 0.0007564102564102564, | |
| "loss": 2.7686, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0007435897435897436, | |
| "loss": 2.7274, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "learning_rate": 0.0007307692307692307, | |
| "loss": 2.7514, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 0.000717948717948718, | |
| "loss": 2.7232, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.0007051282051282052, | |
| "loss": 2.6937, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.0006923076923076923, | |
| "loss": 2.7209, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 0.0006794871794871796, | |
| "loss": 2.6832, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.0006666666666666666, | |
| "loss": 2.708, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.561117649078369, | |
| "eval_runtime": 41.0726, | |
| "eval_samples_per_second": 253.235, | |
| "eval_steps_per_second": 7.937, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "learning_rate": 0.0006538461538461538, | |
| "loss": 2.2119, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 0.0006410256410256411, | |
| "loss": 2.2279, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 0.0006282051282051282, | |
| "loss": 2.1996, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "learning_rate": 0.0006153846153846154, | |
| "loss": 2.1886, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "learning_rate": 0.0006025641025641026, | |
| "loss": 2.268, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0005897435897435898, | |
| "loss": 2.2118, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 0.0005769230769230769, | |
| "loss": 2.2831, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "learning_rate": 0.0005641025641025641, | |
| "loss": 2.2317, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "learning_rate": 0.0005512820512820514, | |
| "loss": 2.2504, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 0.0005384615384615384, | |
| "loss": 2.241, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.0005256410256410256, | |
| "loss": 2.2381, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 0.0005128205128205128, | |
| "loss": 2.2313, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 0.0005, | |
| "loss": 2.2493, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 0.0004871794871794872, | |
| "loss": 2.2056, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 0.00047435897435897434, | |
| "loss": 2.2178, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "learning_rate": 0.0004615384615384616, | |
| "loss": 2.2518, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "learning_rate": 0.0004487179487179487, | |
| "loss": 2.2354, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 0.0004358974358974359, | |
| "loss": 2.2144, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "learning_rate": 0.0004230769230769231, | |
| "loss": 2.2197, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 0.00041025641025641023, | |
| "loss": 2.2006, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "learning_rate": 0.0003974358974358974, | |
| "loss": 2.1802, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 0.00038461538461538467, | |
| "loss": 2.1656, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "learning_rate": 0.0003717948717948718, | |
| "loss": 2.1592, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "learning_rate": 0.000358974358974359, | |
| "loss": 2.1847, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "learning_rate": 0.00034615384615384613, | |
| "loss": 2.1981, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0003333333333333333, | |
| "loss": 2.1768, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.41491436958313, | |
| "eval_runtime": 39.8227, | |
| "eval_samples_per_second": 261.183, | |
| "eval_steps_per_second": 8.186, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "learning_rate": 0.00032051282051282057, | |
| "loss": 1.7593, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "learning_rate": 0.0003076923076923077, | |
| "loss": 1.7591, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.0002948717948717949, | |
| "loss": 1.7473, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 2.15, | |
| "learning_rate": 0.00028205128205128203, | |
| "loss": 1.7836, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.19, | |
| "learning_rate": 0.0002692307692307692, | |
| "loss": 1.7636, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 2.23, | |
| "learning_rate": 0.0002564102564102564, | |
| "loss": 1.763, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.27, | |
| "learning_rate": 0.0002435897435897436, | |
| "loss": 1.7639, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 2.31, | |
| "learning_rate": 0.0002307692307692308, | |
| "loss": 1.791, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.35, | |
| "learning_rate": 0.00021794871794871795, | |
| "loss": 1.7802, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 2.38, | |
| "learning_rate": 0.00020512820512820512, | |
| "loss": 1.7899, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.42, | |
| "learning_rate": 0.00019230769230769233, | |
| "loss": 1.7601, | |
| "step": 3150 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 0.0001794871794871795, | |
| "loss": 1.7696, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 0.00016666666666666666, | |
| "loss": 1.7586, | |
| "step": 3250 | |
| }, | |
| { | |
| "epoch": 2.54, | |
| "learning_rate": 0.00015384615384615385, | |
| "loss": 1.7312, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 0.00014102564102564101, | |
| "loss": 1.7353, | |
| "step": 3350 | |
| }, | |
| { | |
| "epoch": 2.61, | |
| "learning_rate": 0.0001282051282051282, | |
| "loss": 1.7453, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.65, | |
| "learning_rate": 0.0001153846153846154, | |
| "loss": 1.7397, | |
| "step": 3450 | |
| }, | |
| { | |
| "epoch": 2.69, | |
| "learning_rate": 0.00010256410256410256, | |
| "loss": 1.7529, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 8.974358974358975e-05, | |
| "loss": 1.7464, | |
| "step": 3550 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 1.7271, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 6.41025641025641e-05, | |
| "loss": 1.7631, | |
| "step": 3650 | |
| }, | |
| { | |
| "epoch": 2.85, | |
| "learning_rate": 5.128205128205128e-05, | |
| "loss": 1.7462, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "learning_rate": 3.846153846153846e-05, | |
| "loss": 1.7318, | |
| "step": 3750 | |
| }, | |
| { | |
| "epoch": 2.92, | |
| "learning_rate": 2.564102564102564e-05, | |
| "loss": 1.724, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 1.282051282051282e-05, | |
| "loss": 1.7065, | |
| "step": 3850 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.0, | |
| "loss": 1.7189, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.406733751296997, | |
| "eval_runtime": 39.7653, | |
| "eval_samples_per_second": 261.56, | |
| "eval_steps_per_second": 8.198, | |
| "step": 3900 | |
| } | |
| ], | |
| "logging_steps": 50, | |
| "max_steps": 3900, | |
| "num_train_epochs": 3, | |
| "save_steps": 500, | |
| "total_flos": 6647135438315520.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |