| { |
| "best_metric": 0.0093231201171875, |
| "best_model_checkpoint": "model_fewrel_1_8-task9/checkpoint-1890", |
| "epoch": 10.0, |
| "eval_steps": 500, |
| "global_step": 2100, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "eval_loss": 0.012847900390625, |
| "eval_rouge1": 98.6287, |
| "eval_rouge2": 98.0199, |
| "eval_rougeL": 98.3895, |
| "eval_rougeLsum": 98.5995, |
| "eval_runtime": 34.0174, |
| "eval_samples_per_second": 32.924, |
| "eval_steps_per_second": 1.029, |
| "step": 210 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.0115509033203125, |
| "eval_rouge1": 98.9154, |
| "eval_rouge2": 98.5367, |
| "eval_rougeL": 98.8038, |
| "eval_rougeLsum": 98.9131, |
| "eval_runtime": 34.4471, |
| "eval_samples_per_second": 32.514, |
| "eval_steps_per_second": 1.016, |
| "step": 420 |
| }, |
| { |
| "epoch": 2.380952380952381, |
| "grad_norm": 0.4975210428237915, |
| "learning_rate": 0.0008665259359149131, |
| "loss": 0.055, |
| "step": 500 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.01337432861328125, |
| "eval_rouge1": 98.6194, |
| "eval_rouge2": 98.0506, |
| "eval_rougeL": 98.3926, |
| "eval_rougeLsum": 98.5818, |
| "eval_runtime": 33.5087, |
| "eval_samples_per_second": 33.424, |
| "eval_steps_per_second": 1.045, |
| "step": 630 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.01132965087890625, |
| "eval_rouge1": 98.8522, |
| "eval_rouge2": 98.4253, |
| "eval_rougeL": 98.7051, |
| "eval_rougeLsum": 98.8197, |
| "eval_runtime": 33.1182, |
| "eval_samples_per_second": 33.818, |
| "eval_steps_per_second": 1.057, |
| "step": 840 |
| }, |
| { |
| "epoch": 4.761904761904762, |
| "grad_norm": 0.9489365816116333, |
| "learning_rate": 0.0005373650467932121, |
| "loss": 0.0088, |
| "step": 1000 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.01326751708984375, |
| "eval_rouge1": 99.2134, |
| "eval_rouge2": 98.8765, |
| "eval_rougeL": 99.0941, |
| "eval_rougeLsum": 99.2096, |
| "eval_runtime": 36.9437, |
| "eval_samples_per_second": 30.316, |
| "eval_steps_per_second": 0.947, |
| "step": 1050 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.01346588134765625, |
| "eval_rouge1": 99.2312, |
| "eval_rouge2": 98.8839, |
| "eval_rougeL": 99.0944, |
| "eval_rougeLsum": 99.2153, |
| "eval_runtime": 34.9945, |
| "eval_samples_per_second": 32.005, |
| "eval_steps_per_second": 1.0, |
| "step": 1260 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.00937652587890625, |
| "eval_rouge1": 99.5998, |
| "eval_rouge2": 99.3899, |
| "eval_rougeL": 99.5205, |
| "eval_rougeLsum": 99.5994, |
| "eval_runtime": 34.6991, |
| "eval_samples_per_second": 32.278, |
| "eval_steps_per_second": 1.009, |
| "step": 1470 |
| }, |
| { |
| "epoch": 7.142857142857143, |
| "grad_norm": 0.030887478962540627, |
| "learning_rate": 0.00018825509907063325, |
| "loss": 0.0032, |
| "step": 1500 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.010223388671875, |
| "eval_rouge1": 99.4127, |
| "eval_rouge2": 99.1295, |
| "eval_rougeL": 99.3072, |
| "eval_rougeLsum": 99.3986, |
| "eval_runtime": 34.3802, |
| "eval_samples_per_second": 32.577, |
| "eval_steps_per_second": 1.018, |
| "step": 1680 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.0093231201171875, |
| "eval_rouge1": 99.4127, |
| "eval_rouge2": 99.1295, |
| "eval_rougeL": 99.3072, |
| "eval_rougeLsum": 99.3986, |
| "eval_runtime": 34.2644, |
| "eval_samples_per_second": 32.687, |
| "eval_steps_per_second": 1.021, |
| "step": 1890 |
| }, |
| { |
| "epoch": 9.523809523809524, |
| "grad_norm": 0.0026726792566478252, |
| "learning_rate": 5.5845868874357386e-06, |
| "loss": 0.0011, |
| "step": 2000 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.00940704345703125, |
| "eval_rouge1": 99.4127, |
| "eval_rouge2": 99.1295, |
| "eval_rougeL": 99.3072, |
| "eval_rougeLsum": 99.3986, |
| "eval_runtime": 34.3879, |
| "eval_samples_per_second": 32.57, |
| "eval_steps_per_second": 1.018, |
| "step": 2100 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 2100, |
| "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": 2.3099168784384e+16, |
| "train_batch_size": 16, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
