| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.7561820571018995, | |
| "global_step": 3165, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "learning_rate": 1.978494623655914e-05, | |
| "loss": 1.0308, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "learning_rate": 1.954599761051374e-05, | |
| "loss": 0.7656, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "learning_rate": 1.930704898446834e-05, | |
| "loss": 0.8032, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "learning_rate": 1.906810035842294e-05, | |
| "loss": 0.7398, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 1.882915173237754e-05, | |
| "loss": 0.7004, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_accuracy": 0.6535433070866141, | |
| "eval_loss": 0.7174085378646851, | |
| "eval_runtime": 431.9798, | |
| "eval_samples_per_second": 2.058, | |
| "eval_steps_per_second": 2.058, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 1.859020310633214e-05, | |
| "loss": 0.8009, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "learning_rate": 1.835125448028674e-05, | |
| "loss": 0.6551, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "learning_rate": 1.811230585424134e-05, | |
| "loss": 0.6348, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "learning_rate": 1.787335722819594e-05, | |
| "loss": 0.5236, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "learning_rate": 1.763440860215054e-05, | |
| "loss": 0.6735, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "eval_accuracy": 0.7007874015748031, | |
| "eval_loss": 0.6178467869758606, | |
| "eval_runtime": 430.5762, | |
| "eval_samples_per_second": 2.065, | |
| "eval_steps_per_second": 2.065, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 1.7395459976105136e-05, | |
| "loss": 0.7371, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "learning_rate": 1.7156511350059738e-05, | |
| "loss": 0.6957, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.691756272401434e-05, | |
| "loss": 0.592, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "learning_rate": 1.6678614097968937e-05, | |
| "loss": 0.5799, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 1.6439665471923538e-05, | |
| "loss": 0.6165, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "eval_accuracy": 0.7424071991001124, | |
| "eval_loss": 0.5588846206665039, | |
| "eval_runtime": 431.475, | |
| "eval_samples_per_second": 2.06, | |
| "eval_steps_per_second": 2.06, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 1.6200716845878136e-05, | |
| "loss": 0.5045, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 1.5961768219832737e-05, | |
| "loss": 0.6213, | |
| "step": 850 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 1.5722819593787338e-05, | |
| "loss": 0.6248, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "learning_rate": 1.5483870967741936e-05, | |
| "loss": 0.4998, | |
| "step": 950 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "learning_rate": 1.5244922341696537e-05, | |
| "loss": 0.6603, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_accuracy": 0.7559055118110236, | |
| "eval_loss": 0.5710476636886597, | |
| "eval_runtime": 432.0116, | |
| "eval_samples_per_second": 2.058, | |
| "eval_steps_per_second": 2.058, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 1.5005973715651137e-05, | |
| "loss": 0.5468, | |
| "step": 1050 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "learning_rate": 1.4767025089605736e-05, | |
| "loss": 0.511, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 1.4528076463560337e-05, | |
| "loss": 0.5575, | |
| "step": 1150 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "learning_rate": 1.4289127837514935e-05, | |
| "loss": 0.7658, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "learning_rate": 1.4050179211469535e-05, | |
| "loss": 0.5575, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "eval_accuracy": 0.7570303712035995, | |
| "eval_loss": 0.5421488881111145, | |
| "eval_runtime": 430.448, | |
| "eval_samples_per_second": 2.065, | |
| "eval_steps_per_second": 2.065, | |
| "step": 1250 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 1.3811230585424136e-05, | |
| "loss": 0.6252, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 1.3572281959378735e-05, | |
| "loss": 0.4673, | |
| "step": 1350 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.5266, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 1.3094384707287935e-05, | |
| "loss": 0.6353, | |
| "step": 1450 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "learning_rate": 1.2855436081242534e-05, | |
| "loss": 0.589, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "eval_accuracy": 0.750281214848144, | |
| "eval_loss": 0.5329739451408386, | |
| "eval_runtime": 429.8372, | |
| "eval_samples_per_second": 2.068, | |
| "eval_steps_per_second": 2.068, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "learning_rate": 1.2616487455197134e-05, | |
| "loss": 0.6944, | |
| "step": 1550 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 1.2377538829151735e-05, | |
| "loss": 0.4992, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "learning_rate": 1.2138590203106333e-05, | |
| "loss": 0.6429, | |
| "step": 1650 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 1.1899641577060932e-05, | |
| "loss": 0.5243, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "learning_rate": 1.1660692951015533e-05, | |
| "loss": 0.5644, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "eval_accuracy": 0.7559055118110236, | |
| "eval_loss": 0.5390347838401794, | |
| "eval_runtime": 429.8773, | |
| "eval_samples_per_second": 2.068, | |
| "eval_steps_per_second": 2.068, | |
| "step": 1750 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 1.1421744324970133e-05, | |
| "loss": 0.6321, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 1.118279569892473e-05, | |
| "loss": 0.5939, | |
| "step": 1850 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "learning_rate": 1.0943847072879332e-05, | |
| "loss": 0.5477, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 1.0704898446833931e-05, | |
| "loss": 0.5171, | |
| "step": 1950 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 1.0465949820788533e-05, | |
| "loss": 0.503, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_accuracy": 0.7592800899887514, | |
| "eval_loss": 0.5502843856811523, | |
| "eval_runtime": 429.7289, | |
| "eval_samples_per_second": 2.069, | |
| "eval_steps_per_second": 2.069, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 1.0227001194743132e-05, | |
| "loss": 0.6059, | |
| "step": 2050 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 9.98805256869773e-06, | |
| "loss": 0.5716, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "learning_rate": 9.749103942652331e-06, | |
| "loss": 0.5367, | |
| "step": 2150 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 9.51015531660693e-06, | |
| "loss": 0.578, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 9.27120669056153e-06, | |
| "loss": 0.6361, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_accuracy": 0.7637795275590551, | |
| "eval_loss": 0.5347180366516113, | |
| "eval_runtime": 432.0598, | |
| "eval_samples_per_second": 2.058, | |
| "eval_steps_per_second": 2.058, | |
| "step": 2250 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "learning_rate": 9.03225806451613e-06, | |
| "loss": 0.5387, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "learning_rate": 8.793309438470729e-06, | |
| "loss": 0.5409, | |
| "step": 2350 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 8.55436081242533e-06, | |
| "loss": 0.4895, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 8.315412186379928e-06, | |
| "loss": 0.5966, | |
| "step": 2450 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 8.07646356033453e-06, | |
| "loss": 0.517, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.7649043869516311, | |
| "eval_loss": 0.5409161448478699, | |
| "eval_runtime": 432.0753, | |
| "eval_samples_per_second": 2.058, | |
| "eval_steps_per_second": 2.058, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "learning_rate": 7.837514934289129e-06, | |
| "loss": 0.5852, | |
| "step": 2550 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 7.5985663082437275e-06, | |
| "loss": 0.5853, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 7.359617682198328e-06, | |
| "loss": 0.6096, | |
| "step": 2650 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "learning_rate": 7.120669056152928e-06, | |
| "loss": 0.6285, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "learning_rate": 6.881720430107528e-06, | |
| "loss": 0.5481, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "eval_accuracy": 0.7525309336332958, | |
| "eval_loss": 0.5184136033058167, | |
| "eval_runtime": 431.9074, | |
| "eval_samples_per_second": 2.058, | |
| "eval_steps_per_second": 2.058, | |
| "step": 2750 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 6.642771804062127e-06, | |
| "loss": 0.6294, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "learning_rate": 6.403823178016727e-06, | |
| "loss": 0.5585, | |
| "step": 2850 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 6.164874551971327e-06, | |
| "loss": 0.5914, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 5.925925925925926e-06, | |
| "loss": 0.5831, | |
| "step": 2950 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "learning_rate": 5.686977299880526e-06, | |
| "loss": 0.6036, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "eval_accuracy": 0.7592800899887514, | |
| "eval_loss": 0.5061925649642944, | |
| "eval_runtime": 431.9996, | |
| "eval_samples_per_second": 2.058, | |
| "eval_steps_per_second": 2.058, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 5.4480286738351265e-06, | |
| "loss": 0.4684, | |
| "step": 3050 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "learning_rate": 5.209080047789725e-06, | |
| "loss": 0.5459, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "learning_rate": 4.9701314217443256e-06, | |
| "loss": 0.5358, | |
| "step": 3150 | |
| } | |
| ], | |
| "max_steps": 4185, | |
| "num_train_epochs": 1, | |
| "total_flos": 0.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |