| { |
| "best_metric": 0.0300140380859375, |
| "best_model_checkpoint": "model_fewrel_1_1-task2/checkpoint-1260", |
| "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": 0.0692138671875, |
| "eval_rouge1": 95.7517, |
| "eval_rouge2": 94.6841, |
| "eval_rougeL": 95.6971, |
| "eval_rougeLsum": 95.7331, |
| "eval_runtime": 33.2038, |
| "eval_samples_per_second": 33.731, |
| "eval_steps_per_second": 1.054, |
| "step": 210 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 0.034423828125, |
| "eval_rouge1": 97.351, |
| "eval_rouge2": 96.6251, |
| "eval_rougeL": 97.3032, |
| "eval_rougeLsum": 97.2964, |
| "eval_runtime": 32.6308, |
| "eval_samples_per_second": 34.323, |
| "eval_steps_per_second": 1.073, |
| "step": 420 |
| }, |
| { |
| "epoch": 2.380952380952381, |
| "grad_norm": 0.4508669972419739, |
| "learning_rate": 0.0008665259359149131, |
| "loss": 0.0924, |
| "step": 500 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 0.03924560546875, |
| "eval_rouge1": 97.2483, |
| "eval_rouge2": 96.6208, |
| "eval_rougeL": 97.2291, |
| "eval_rougeLsum": 97.2002, |
| "eval_runtime": 33.4876, |
| "eval_samples_per_second": 33.445, |
| "eval_steps_per_second": 1.045, |
| "step": 630 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_loss": 0.0413818359375, |
| "eval_rouge1": 97.1906, |
| "eval_rouge2": 96.559, |
| "eval_rougeL": 97.1839, |
| "eval_rougeLsum": 97.1621, |
| "eval_runtime": 34.4429, |
| "eval_samples_per_second": 32.518, |
| "eval_steps_per_second": 1.016, |
| "step": 840 |
| }, |
| { |
| "epoch": 4.761904761904762, |
| "grad_norm": 0.1619143784046173, |
| "learning_rate": 0.0005373650467932121, |
| "loss": 0.0276, |
| "step": 1000 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_loss": 0.042449951171875, |
| "eval_rouge1": 97.4464, |
| "eval_rouge2": 96.8611, |
| "eval_rougeL": 97.4297, |
| "eval_rougeLsum": 97.4266, |
| "eval_runtime": 32.4102, |
| "eval_samples_per_second": 34.557, |
| "eval_steps_per_second": 1.08, |
| "step": 1050 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_loss": 0.0300140380859375, |
| "eval_rouge1": 98.1516, |
| "eval_rouge2": 97.6994, |
| "eval_rougeL": 98.1475, |
| "eval_rougeLsum": 98.155, |
| "eval_runtime": 32.3626, |
| "eval_samples_per_second": 34.608, |
| "eval_steps_per_second": 1.081, |
| "step": 1260 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_loss": 0.031494140625, |
| "eval_rouge1": 97.6953, |
| "eval_rouge2": 97.1861, |
| "eval_rougeL": 97.7355, |
| "eval_rougeLsum": 97.713, |
| "eval_runtime": 31.6892, |
| "eval_samples_per_second": 35.343, |
| "eval_steps_per_second": 1.104, |
| "step": 1470 |
| }, |
| { |
| "epoch": 7.142857142857143, |
| "grad_norm": 0.17737896740436554, |
| "learning_rate": 0.00018825509907063325, |
| "loss": 0.0138, |
| "step": 1500 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_loss": 0.03729248046875, |
| "eval_rouge1": 98.0557, |
| "eval_rouge2": 97.5844, |
| "eval_rougeL": 98.0434, |
| "eval_rougeLsum": 98.0396, |
| "eval_runtime": 32.4888, |
| "eval_samples_per_second": 34.473, |
| "eval_steps_per_second": 1.077, |
| "step": 1680 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_loss": 0.0333251953125, |
| "eval_rouge1": 98.2547, |
| "eval_rouge2": 97.8119, |
| "eval_rougeL": 98.2452, |
| "eval_rougeLsum": 98.2669, |
| "eval_runtime": 31.823, |
| "eval_samples_per_second": 35.195, |
| "eval_steps_per_second": 1.1, |
| "step": 1890 |
| }, |
| { |
| "epoch": 9.523809523809524, |
| "grad_norm": 0.1923867166042328, |
| "learning_rate": 5.5845868874357386e-06, |
| "loss": 0.0088, |
| "step": 2000 |
| }, |
| { |
| "epoch": 10.0, |
| "eval_loss": 0.03363037109375, |
| "eval_rouge1": 98.2547, |
| "eval_rouge2": 97.8119, |
| "eval_rougeL": 98.2452, |
| "eval_rougeLsum": 98.2669, |
| "eval_runtime": 31.6927, |
| "eval_samples_per_second": 35.339, |
| "eval_steps_per_second": 1.104, |
| "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 |
| } |
|
|
