| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.003338898163606, | |
| "eval_steps": 10, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.07, | |
| "learning_rate": 3e-05, | |
| "loss": 1.5245, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_accuracy": 0.3355263157894737, | |
| "eval_loss": 0.6506821513175964, | |
| "eval_runtime": 16.9243, | |
| "eval_samples_per_second": 17.962, | |
| "eval_steps_per_second": 4.491, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6666, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "eval_accuracy": 0.3815789473684211, | |
| "eval_loss": 0.6464425325393677, | |
| "eval_runtime": 16.9195, | |
| "eval_samples_per_second": 17.967, | |
| "eval_steps_per_second": 4.492, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6527, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_accuracy": 0.3684210526315789, | |
| "eval_loss": 0.6426967978477478, | |
| "eval_runtime": 16.9282, | |
| "eval_samples_per_second": 17.958, | |
| "eval_steps_per_second": 4.49, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6168, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_accuracy": 0.3980263157894737, | |
| "eval_loss": 0.6321499943733215, | |
| "eval_runtime": 17.0092, | |
| "eval_samples_per_second": 17.873, | |
| "eval_steps_per_second": 4.468, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6584, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "eval_accuracy": 0.39144736842105265, | |
| "eval_loss": 0.6181844472885132, | |
| "eval_runtime": 16.9419, | |
| "eval_samples_per_second": 17.944, | |
| "eval_steps_per_second": 4.486, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "learning_rate": 3e-05, | |
| "loss": 0.586, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "eval_accuracy": 0.4144736842105263, | |
| "eval_loss": 0.6244160532951355, | |
| "eval_runtime": 16.9269, | |
| "eval_samples_per_second": 17.96, | |
| "eval_steps_per_second": 4.49, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5924, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "eval_accuracy": 0.4342105263157895, | |
| "eval_loss": 0.6033625602722168, | |
| "eval_runtime": 16.9236, | |
| "eval_samples_per_second": 17.963, | |
| "eval_steps_per_second": 4.491, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6069, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "eval_accuracy": 0.4375, | |
| "eval_loss": 0.6096391677856445, | |
| "eval_runtime": 16.9238, | |
| "eval_samples_per_second": 17.963, | |
| "eval_steps_per_second": 4.491, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5999, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "eval_accuracy": 0.4407894736842105, | |
| "eval_loss": 0.6095999479293823, | |
| "eval_runtime": 16.9342, | |
| "eval_samples_per_second": 17.952, | |
| "eval_steps_per_second": 4.488, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6206, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "eval_accuracy": 0.45723684210526316, | |
| "eval_loss": 0.607021152973175, | |
| "eval_runtime": 16.9304, | |
| "eval_samples_per_second": 17.956, | |
| "eval_steps_per_second": 4.489, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5793, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "eval_accuracy": 0.45723684210526316, | |
| "eval_loss": 0.601601243019104, | |
| "eval_runtime": 16.9375, | |
| "eval_samples_per_second": 17.948, | |
| "eval_steps_per_second": 4.487, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 3e-05, | |
| "loss": 0.6208, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "eval_accuracy": 0.4605263157894737, | |
| "eval_loss": 0.5902404189109802, | |
| "eval_runtime": 16.924, | |
| "eval_samples_per_second": 17.963, | |
| "eval_steps_per_second": 4.491, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5622, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "eval_accuracy": 0.4769736842105263, | |
| "eval_loss": 0.5775408744812012, | |
| "eval_runtime": 16.9329, | |
| "eval_samples_per_second": 17.953, | |
| "eval_steps_per_second": 4.488, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5502, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "eval_accuracy": 0.46710526315789475, | |
| "eval_loss": 0.57607102394104, | |
| "eval_runtime": 16.9226, | |
| "eval_samples_per_second": 17.964, | |
| "eval_steps_per_second": 4.491, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5958, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.4901315789473684, | |
| "eval_loss": 0.5606401562690735, | |
| "eval_runtime": 16.929, | |
| "eval_samples_per_second": 17.957, | |
| "eval_steps_per_second": 4.489, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4558, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "eval_accuracy": 0.47368421052631576, | |
| "eval_loss": 0.5839833617210388, | |
| "eval_runtime": 16.9304, | |
| "eval_samples_per_second": 17.956, | |
| "eval_steps_per_second": 4.489, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4411, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "eval_accuracy": 0.4901315789473684, | |
| "eval_loss": 0.5631235837936401, | |
| "eval_runtime": 16.9238, | |
| "eval_samples_per_second": 17.963, | |
| "eval_steps_per_second": 4.491, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4144, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 0.5744868516921997, | |
| "eval_runtime": 16.9382, | |
| "eval_samples_per_second": 17.948, | |
| "eval_steps_per_second": 4.487, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4647, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "eval_accuracy": 0.4605263157894737, | |
| "eval_loss": 0.593177080154419, | |
| "eval_runtime": 16.932, | |
| "eval_samples_per_second": 17.954, | |
| "eval_steps_per_second": 4.489, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4504, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "eval_accuracy": 0.5098684210526315, | |
| "eval_loss": 0.5798581838607788, | |
| "eval_runtime": 16.9337, | |
| "eval_samples_per_second": 17.952, | |
| "eval_steps_per_second": 4.488, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4299, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "eval_accuracy": 0.4934210526315789, | |
| "eval_loss": 0.64882493019104, | |
| "eval_runtime": 16.9391, | |
| "eval_samples_per_second": 17.947, | |
| "eval_steps_per_second": 4.487, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "learning_rate": 3e-05, | |
| "loss": 0.425, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "eval_accuracy": 0.5131578947368421, | |
| "eval_loss": 0.5704348683357239, | |
| "eval_runtime": 16.9325, | |
| "eval_samples_per_second": 17.954, | |
| "eval_steps_per_second": 4.488, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4152, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_accuracy": 0.506578947368421, | |
| "eval_loss": 0.5582014322280884, | |
| "eval_runtime": 16.9258, | |
| "eval_samples_per_second": 17.961, | |
| "eval_steps_per_second": 4.49, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 3e-05, | |
| "loss": 0.425, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_accuracy": 0.5328947368421053, | |
| "eval_loss": 0.5488855838775635, | |
| "eval_runtime": 16.9288, | |
| "eval_samples_per_second": 17.958, | |
| "eval_steps_per_second": 4.489, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "learning_rate": 3e-05, | |
| "loss": 0.446, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "eval_accuracy": 0.5197368421052632, | |
| "eval_loss": 0.5479023456573486, | |
| "eval_runtime": 16.9319, | |
| "eval_samples_per_second": 17.954, | |
| "eval_steps_per_second": 4.489, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3908, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "eval_accuracy": 0.5164473684210527, | |
| "eval_loss": 0.5564107894897461, | |
| "eval_runtime": 16.9414, | |
| "eval_samples_per_second": 17.944, | |
| "eval_steps_per_second": 4.486, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "learning_rate": 3e-05, | |
| "loss": 0.443, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_accuracy": 0.5032894736842105, | |
| "eval_loss": 0.5418796539306641, | |
| "eval_runtime": 16.9208, | |
| "eval_samples_per_second": 17.966, | |
| "eval_steps_per_second": 4.492, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4081, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_accuracy": 0.506578947368421, | |
| "eval_loss": 0.5948407053947449, | |
| "eval_runtime": 16.9289, | |
| "eval_samples_per_second": 17.957, | |
| "eval_steps_per_second": 4.489, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "learning_rate": 3e-05, | |
| "loss": 0.3944, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.5394736842105263, | |
| "eval_loss": 0.554680347442627, | |
| "eval_runtime": 16.9311, | |
| "eval_samples_per_second": 17.955, | |
| "eval_steps_per_second": 4.489, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.4005, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5361842105263158, | |
| "eval_loss": 0.5615983009338379, | |
| "eval_runtime": 16.9277, | |
| "eval_samples_per_second": 17.959, | |
| "eval_steps_per_second": 4.49, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 300, | |
| "total_flos": 7.380159778455552e+16, | |
| "train_loss": 0.5479170862833659, | |
| "train_runtime": 1462.1781, | |
| "train_samples_per_second": 3.283, | |
| "train_steps_per_second": 0.205 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 150, | |
| "total_flos": 7.380159778455552e+16, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |