| { | |
| "best_metric": 0.518923282623291, | |
| "best_model_checkpoint": "./beans_outputs/checkpoint-1300", | |
| "epoch": 10.0, | |
| "eval_steps": 500, | |
| "global_step": 1300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.7692307692307693, | |
| "grad_norm": 5.921261310577393, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1031, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.5037593984962406, | |
| "eval_loss": 1.031303882598877, | |
| "eval_runtime": 1.1329, | |
| "eval_samples_per_second": 117.399, | |
| "eval_steps_per_second": 15.006, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.5384615384615383, | |
| "grad_norm": 2.5376038551330566, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0087, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5789473684210527, | |
| "eval_loss": 0.9253108501434326, | |
| "eval_runtime": 0.9794, | |
| "eval_samples_per_second": 135.804, | |
| "eval_steps_per_second": 17.358, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.3076923076923075, | |
| "grad_norm": 4.594298839569092, | |
| "learning_rate": 5e-05, | |
| "loss": 0.8781, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6616541353383458, | |
| "eval_loss": 0.8822579979896545, | |
| "eval_runtime": 0.9528, | |
| "eval_samples_per_second": 139.584, | |
| "eval_steps_per_second": 17.842, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 9.547410011291504, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7773, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 3.8461538461538463, | |
| "grad_norm": 2.8391215801239014, | |
| "learning_rate": 5e-05, | |
| "loss": 0.7127, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.706766917293233, | |
| "eval_loss": 0.6853336691856384, | |
| "eval_runtime": 1.1565, | |
| "eval_samples_per_second": 114.999, | |
| "eval_steps_per_second": 14.699, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 4.615384615384615, | |
| "grad_norm": 4.75039005279541, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6784, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7142857142857143, | |
| "eval_loss": 0.7131139039993286, | |
| "eval_runtime": 1.1416, | |
| "eval_samples_per_second": 116.5, | |
| "eval_steps_per_second": 14.891, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 5.384615384615385, | |
| "grad_norm": 6.33519983291626, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6864, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6992481203007519, | |
| "eval_loss": 0.7313814759254456, | |
| "eval_runtime": 1.0995, | |
| "eval_samples_per_second": 120.962, | |
| "eval_steps_per_second": 15.461, | |
| "step": 780 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 5.234774589538574, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6133, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 6.923076923076923, | |
| "grad_norm": 17.76725196838379, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5986, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7218045112781954, | |
| "eval_loss": 0.6223586201667786, | |
| "eval_runtime": 0.9687, | |
| "eval_samples_per_second": 137.298, | |
| "eval_steps_per_second": 17.549, | |
| "step": 910 | |
| }, | |
| { | |
| "epoch": 7.6923076923076925, | |
| "grad_norm": 4.8591203689575195, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5956, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7744360902255639, | |
| "eval_loss": 0.526090145111084, | |
| "eval_runtime": 1.1771, | |
| "eval_samples_per_second": 112.993, | |
| "eval_steps_per_second": 14.443, | |
| "step": 1040 | |
| }, | |
| { | |
| "epoch": 8.461538461538462, | |
| "grad_norm": 8.905884742736816, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6009, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8120300751879699, | |
| "eval_loss": 0.5274345874786377, | |
| "eval_runtime": 1.0432, | |
| "eval_samples_per_second": 127.497, | |
| "eval_steps_per_second": 16.297, | |
| "step": 1170 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 7.226690292358398, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6035, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 51.65995407104492, | |
| "learning_rate": 5e-05, | |
| "loss": 0.5433, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8045112781954887, | |
| "eval_loss": 0.518923282623291, | |
| "eval_runtime": 1.0446, | |
| "eval_samples_per_second": 127.324, | |
| "eval_steps_per_second": 16.275, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 1300, | |
| "total_flos": 8.400578669044531e+17, | |
| "train_loss": 0.7230858435997596, | |
| "train_runtime": 200.7702, | |
| "train_samples_per_second": 51.502, | |
| "train_steps_per_second": 6.475 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 1300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.400578669044531e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |