| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.6298815822625347, | |
| "eval_steps": 500, | |
| "global_step": 10000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03, | |
| "learning_rate": 9.5e-05, | |
| "loss": 37.4397, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_loss": 31.211898803710938, | |
| "eval_runtime": 6.944, | |
| "eval_samples_per_second": 14.401, | |
| "eval_steps_per_second": 3.6, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "learning_rate": 9e-05, | |
| "loss": 33.7011, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "eval_loss": 30.628690719604492, | |
| "eval_runtime": 8.8362, | |
| "eval_samples_per_second": 11.317, | |
| "eval_steps_per_second": 2.829, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "learning_rate": 8.5e-05, | |
| "loss": 34.3949, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "eval_loss": 30.2351016998291, | |
| "eval_runtime": 9.7911, | |
| "eval_samples_per_second": 10.213, | |
| "eval_steps_per_second": 2.553, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 8e-05, | |
| "loss": 33.7534, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_loss": 30.151206970214844, | |
| "eval_runtime": 9.0893, | |
| "eval_samples_per_second": 11.002, | |
| "eval_steps_per_second": 2.75, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 33.9418, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "eval_loss": 30.117929458618164, | |
| "eval_runtime": 7.9747, | |
| "eval_samples_per_second": 12.54, | |
| "eval_steps_per_second": 3.135, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "learning_rate": 7e-05, | |
| "loss": 33.3573, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "eval_loss": 29.591096878051758, | |
| "eval_runtime": 7.5459, | |
| "eval_samples_per_second": 13.252, | |
| "eval_steps_per_second": 3.313, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 33.3372, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "eval_loss": 29.37584686279297, | |
| "eval_runtime": 9.1309, | |
| "eval_samples_per_second": 10.952, | |
| "eval_steps_per_second": 2.738, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "learning_rate": 6e-05, | |
| "loss": 32.5095, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "eval_loss": 29.21722412109375, | |
| "eval_runtime": 8.935, | |
| "eval_samples_per_second": 11.192, | |
| "eval_steps_per_second": 2.798, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 32.6192, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "eval_loss": 28.85555648803711, | |
| "eval_runtime": 8.0464, | |
| "eval_samples_per_second": 12.428, | |
| "eval_steps_per_second": 3.107, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 5e-05, | |
| "loss": 32.902, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_loss": 28.798521041870117, | |
| "eval_runtime": 8.3773, | |
| "eval_samples_per_second": 11.937, | |
| "eval_steps_per_second": 2.984, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 4.5e-05, | |
| "loss": 33.2756, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "eval_loss": 28.811630249023438, | |
| "eval_runtime": 8.4098, | |
| "eval_samples_per_second": 11.891, | |
| "eval_steps_per_second": 2.973, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "learning_rate": 4e-05, | |
| "loss": 32.3762, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "eval_loss": 28.51260757446289, | |
| "eval_runtime": 8.6441, | |
| "eval_samples_per_second": 11.569, | |
| "eval_steps_per_second": 2.892, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "learning_rate": 3.5e-05, | |
| "loss": 32.5452, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_loss": 28.600557327270508, | |
| "eval_runtime": 7.2739, | |
| "eval_samples_per_second": 13.748, | |
| "eval_steps_per_second": 3.437, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "learning_rate": 3e-05, | |
| "loss": 33.0182, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 28.495351791381836, | |
| "eval_runtime": 6.2937, | |
| "eval_samples_per_second": 15.889, | |
| "eval_steps_per_second": 3.972, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 2.5e-05, | |
| "loss": 31.8452, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_loss": 28.361299514770508, | |
| "eval_runtime": 8.0596, | |
| "eval_samples_per_second": 12.407, | |
| "eval_steps_per_second": 3.102, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 2e-05, | |
| "loss": 32.0009, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 28.153642654418945, | |
| "eval_runtime": 6.3145, | |
| "eval_samples_per_second": 15.837, | |
| "eval_steps_per_second": 3.959, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "learning_rate": 1.5e-05, | |
| "loss": 33.4938, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_loss": 28.087358474731445, | |
| "eval_runtime": 6.3508, | |
| "eval_samples_per_second": 15.746, | |
| "eval_steps_per_second": 3.937, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "learning_rate": 1e-05, | |
| "loss": 31.9753, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "eval_loss": 28.113698959350586, | |
| "eval_runtime": 9.8066, | |
| "eval_samples_per_second": 10.197, | |
| "eval_steps_per_second": 2.549, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 5e-06, | |
| "loss": 31.9571, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_loss": 28.042316436767578, | |
| "eval_runtime": 9.4344, | |
| "eval_samples_per_second": 10.6, | |
| "eval_steps_per_second": 2.65, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "learning_rate": 0.0, | |
| "loss": 31.8931, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "eval_loss": 28.017412185668945, | |
| "eval_runtime": 9.3747, | |
| "eval_samples_per_second": 10.667, | |
| "eval_steps_per_second": 2.667, | |
| "step": 10000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10000, | |
| "num_train_epochs": 1, | |
| "save_steps": 5000, | |
| "total_flos": 1.369514704896e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |