| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.2597631645250693, | |
| "eval_steps": 500, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9.5e-05, | |
| "loss": 1.6085, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 1.3112831115722656, | |
| "eval_runtime": 3.3067, | |
| "eval_samples_per_second": 30.241, | |
| "eval_steps_per_second": 3.931, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 9e-05, | |
| "loss": 1.5215, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 1.5054197311401367, | |
| "eval_runtime": 3.1798, | |
| "eval_samples_per_second": 31.449, | |
| "eval_steps_per_second": 4.088, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 8.5e-05, | |
| "loss": 1.4697, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 1.32753324508667, | |
| "eval_runtime": 3.3247, | |
| "eval_samples_per_second": 30.078, | |
| "eval_steps_per_second": 3.91, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 8e-05, | |
| "loss": 1.4948, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 1.4151370525360107, | |
| "eval_runtime": 3.174, | |
| "eval_samples_per_second": 31.506, | |
| "eval_steps_per_second": 4.096, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 1.4787, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 1.1171963214874268, | |
| "eval_runtime": 3.1742, | |
| "eval_samples_per_second": 31.504, | |
| "eval_steps_per_second": 4.096, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 7e-05, | |
| "loss": 1.4196, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_loss": 1.495429277420044, | |
| "eval_runtime": 3.1924, | |
| "eval_samples_per_second": 31.325, | |
| "eval_steps_per_second": 4.072, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 1.4509, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 1.3065623044967651, | |
| "eval_runtime": 3.1668, | |
| "eval_samples_per_second": 31.577, | |
| "eval_steps_per_second": 4.105, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 6e-05, | |
| "loss": 1.4333, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 1.4525870084762573, | |
| "eval_runtime": 3.1817, | |
| "eval_samples_per_second": 31.43, | |
| "eval_steps_per_second": 4.086, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 1.4404, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 1.3187968730926514, | |
| "eval_runtime": 3.191, | |
| "eval_samples_per_second": 31.338, | |
| "eval_steps_per_second": 4.074, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 5e-05, | |
| "loss": 1.4408, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 1.1972750425338745, | |
| "eval_runtime": 3.1767, | |
| "eval_samples_per_second": 31.479, | |
| "eval_steps_per_second": 4.092, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.4296, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "eval_loss": 1.3503276109695435, | |
| "eval_runtime": 3.1871, | |
| "eval_samples_per_second": 31.376, | |
| "eval_steps_per_second": 4.079, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "learning_rate": 4e-05, | |
| "loss": 1.3828, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "eval_loss": 1.1146811246871948, | |
| "eval_runtime": 3.1707, | |
| "eval_samples_per_second": 31.538, | |
| "eval_steps_per_second": 4.1, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.482, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_loss": 1.4396498203277588, | |
| "eval_runtime": 3.2045, | |
| "eval_samples_per_second": 31.206, | |
| "eval_steps_per_second": 4.057, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4022, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_loss": 1.2612046003341675, | |
| "eval_runtime": 3.187, | |
| "eval_samples_per_second": 31.378, | |
| "eval_steps_per_second": 4.079, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.4278, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "eval_loss": 1.3243725299835205, | |
| "eval_runtime": 3.2028, | |
| "eval_samples_per_second": 31.223, | |
| "eval_steps_per_second": 4.059, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "learning_rate": 2e-05, | |
| "loss": 1.3767, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "eval_loss": 1.3397793769836426, | |
| "eval_runtime": 3.2017, | |
| "eval_samples_per_second": 31.234, | |
| "eval_steps_per_second": 4.06, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.346, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_loss": 1.278018593788147, | |
| "eval_runtime": 3.373, | |
| "eval_samples_per_second": 29.647, | |
| "eval_steps_per_second": 3.854, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "learning_rate": 1e-05, | |
| "loss": 1.3513, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "eval_loss": 1.3960318565368652, | |
| "eval_runtime": 3.2292, | |
| "eval_samples_per_second": 30.967, | |
| "eval_steps_per_second": 4.026, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 5e-06, | |
| "loss": 1.3398, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_loss": 1.2023816108703613, | |
| "eval_runtime": 3.2042, | |
| "eval_samples_per_second": 31.209, | |
| "eval_steps_per_second": 4.057, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "learning_rate": 0.0, | |
| "loss": 1.3608, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_loss": 1.2703720331192017, | |
| "eval_runtime": 3.2009, | |
| "eval_samples_per_second": 31.241, | |
| "eval_steps_per_second": 4.061, | |
| "step": 10000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10000, | |
| "num_train_epochs": 2, | |
| "save_steps": 5000, | |
| "total_flos": 1.3694633480945664e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |