| { |
| "best_metric": 42.7449, |
| "best_model_checkpoint": "opus_base_wce_random/checkpoint-80000", |
| "epoch": 3.943217665615142, |
| "global_step": 80000, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.2, |
| "learning_rate": 1.9753918572555207e-05, |
| "loss": 0.2174, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.2, |
| "eval_bleu": 38.1466, |
| "eval_gen_len": 37.2387, |
| "eval_loss": 0.1091395914554596, |
| "eval_runtime": 169.6196, |
| "eval_samples_per_second": 6.149, |
| "eval_steps_per_second": 0.195, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.39, |
| "learning_rate": 1.950746746845426e-05, |
| "loss": 0.1833, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.39, |
| "eval_bleu": 40.8678, |
| "eval_gen_len": 34.2282, |
| "eval_loss": 0.1069067046046257, |
| "eval_runtime": 187.3165, |
| "eval_samples_per_second": 5.568, |
| "eval_steps_per_second": 0.176, |
| "step": 8000 |
| }, |
| { |
| "epoch": 0.59, |
| "learning_rate": 1.9261016364353314e-05, |
| "loss": 0.176, |
| "step": 12000 |
| }, |
| { |
| "epoch": 0.59, |
| "eval_bleu": 41.2373, |
| "eval_gen_len": 34.722, |
| "eval_loss": 0.10556207597255707, |
| "eval_runtime": 176.6792, |
| "eval_samples_per_second": 5.903, |
| "eval_steps_per_second": 0.187, |
| "step": 12000 |
| }, |
| { |
| "epoch": 0.79, |
| "learning_rate": 1.9014565260252367e-05, |
| "loss": 0.1702, |
| "step": 16000 |
| }, |
| { |
| "epoch": 0.79, |
| "eval_bleu": 41.2236, |
| "eval_gen_len": 35.4794, |
| "eval_loss": 0.10456866770982742, |
| "eval_runtime": 126.4686, |
| "eval_samples_per_second": 8.247, |
| "eval_steps_per_second": 0.261, |
| "step": 16000 |
| }, |
| { |
| "epoch": 0.99, |
| "learning_rate": 1.876811415615142e-05, |
| "loss": 0.1677, |
| "step": 20000 |
| }, |
| { |
| "epoch": 0.99, |
| "eval_bleu": 41.6373, |
| "eval_gen_len": 35.6874, |
| "eval_loss": 0.10391870141029358, |
| "eval_runtime": 137.3482, |
| "eval_samples_per_second": 7.594, |
| "eval_steps_per_second": 0.24, |
| "step": 20000 |
| }, |
| { |
| "epoch": 1.18, |
| "learning_rate": 1.8521786277602524e-05, |
| "loss": 0.1556, |
| "step": 24000 |
| }, |
| { |
| "epoch": 1.18, |
| "eval_bleu": 41.2704, |
| "eval_gen_len": 35.7728, |
| "eval_loss": 0.10389306396245956, |
| "eval_runtime": 129.1131, |
| "eval_samples_per_second": 8.078, |
| "eval_steps_per_second": 0.256, |
| "step": 24000 |
| }, |
| { |
| "epoch": 1.38, |
| "learning_rate": 1.8275520011829655e-05, |
| "loss": 0.1553, |
| "step": 28000 |
| }, |
| { |
| "epoch": 1.38, |
| "eval_bleu": 41.6866, |
| "eval_gen_len": 35.348, |
| "eval_loss": 0.10308653861284256, |
| "eval_runtime": 149.5149, |
| "eval_samples_per_second": 6.976, |
| "eval_steps_per_second": 0.221, |
| "step": 28000 |
| }, |
| { |
| "epoch": 1.58, |
| "learning_rate": 1.8029130520504733e-05, |
| "loss": 0.1548, |
| "step": 32000 |
| }, |
| { |
| "epoch": 1.58, |
| "eval_bleu": 41.3994, |
| "eval_gen_len": 35.7287, |
| "eval_loss": 0.10239578038454056, |
| "eval_runtime": 125.8581, |
| "eval_samples_per_second": 8.287, |
| "eval_steps_per_second": 0.262, |
| "step": 32000 |
| }, |
| { |
| "epoch": 1.77, |
| "learning_rate": 1.7782802641955836e-05, |
| "loss": 0.1533, |
| "step": 36000 |
| }, |
| { |
| "epoch": 1.77, |
| "eval_bleu": 39.5845, |
| "eval_gen_len": 39.464, |
| "eval_loss": 0.102129265666008, |
| "eval_runtime": 123.946, |
| "eval_samples_per_second": 8.415, |
| "eval_steps_per_second": 0.266, |
| "step": 36000 |
| }, |
| { |
| "epoch": 1.97, |
| "learning_rate": 1.7536413150630914e-05, |
| "loss": 0.1525, |
| "step": 40000 |
| }, |
| { |
| "epoch": 1.97, |
| "eval_bleu": 42.2251, |
| "eval_gen_len": 34.3845, |
| "eval_loss": 0.10222680866718292, |
| "eval_runtime": 101.5149, |
| "eval_samples_per_second": 10.274, |
| "eval_steps_per_second": 0.325, |
| "step": 40000 |
| }, |
| { |
| "epoch": 2.17, |
| "learning_rate": 1.729008527208202e-05, |
| "loss": 0.1446, |
| "step": 44000 |
| }, |
| { |
| "epoch": 2.17, |
| "eval_bleu": 41.9432, |
| "eval_gen_len": 35.0316, |
| "eval_loss": 0.10267173498868942, |
| "eval_runtime": 129.0759, |
| "eval_samples_per_second": 8.081, |
| "eval_steps_per_second": 0.256, |
| "step": 44000 |
| }, |
| { |
| "epoch": 2.37, |
| "learning_rate": 1.7043757393533123e-05, |
| "loss": 0.1436, |
| "step": 48000 |
| }, |
| { |
| "epoch": 2.37, |
| "eval_bleu": 42.028, |
| "eval_gen_len": 35.8677, |
| "eval_loss": 0.102423757314682, |
| "eval_runtime": 128.7526, |
| "eval_samples_per_second": 8.101, |
| "eval_steps_per_second": 0.256, |
| "step": 48000 |
| }, |
| { |
| "epoch": 2.56, |
| "learning_rate": 1.679742951498423e-05, |
| "loss": 0.1425, |
| "step": 52000 |
| }, |
| { |
| "epoch": 2.56, |
| "eval_bleu": 41.9107, |
| "eval_gen_len": 35.6884, |
| "eval_loss": 0.1020052582025528, |
| "eval_runtime": 98.4633, |
| "eval_samples_per_second": 10.593, |
| "eval_steps_per_second": 0.335, |
| "step": 52000 |
| }, |
| { |
| "epoch": 2.76, |
| "learning_rate": 1.6551040023659308e-05, |
| "loss": 0.1427, |
| "step": 56000 |
| }, |
| { |
| "epoch": 2.76, |
| "eval_bleu": 41.898, |
| "eval_gen_len": 35.7948, |
| "eval_loss": 0.10190638899803162, |
| "eval_runtime": 155.172, |
| "eval_samples_per_second": 6.722, |
| "eval_steps_per_second": 0.213, |
| "step": 56000 |
| }, |
| { |
| "epoch": 2.96, |
| "learning_rate": 1.630471214511041e-05, |
| "loss": 0.1431, |
| "step": 60000 |
| }, |
| { |
| "epoch": 2.96, |
| "eval_bleu": 42.4405, |
| "eval_gen_len": 34.8159, |
| "eval_loss": 0.10122686624526978, |
| "eval_runtime": 109.5432, |
| "eval_samples_per_second": 9.521, |
| "eval_steps_per_second": 0.301, |
| "step": 60000 |
| }, |
| { |
| "epoch": 3.15, |
| "learning_rate": 1.6058384266561514e-05, |
| "loss": 0.1365, |
| "step": 64000 |
| }, |
| { |
| "epoch": 3.15, |
| "eval_bleu": 42.1656, |
| "eval_gen_len": 35.1611, |
| "eval_loss": 0.1020784005522728, |
| "eval_runtime": 127.0613, |
| "eval_samples_per_second": 8.209, |
| "eval_steps_per_second": 0.26, |
| "step": 64000 |
| }, |
| { |
| "epoch": 3.35, |
| "learning_rate": 1.581205638801262e-05, |
| "loss": 0.1359, |
| "step": 68000 |
| }, |
| { |
| "epoch": 3.35, |
| "eval_bleu": 42.2236, |
| "eval_gen_len": 34.6663, |
| "eval_loss": 0.10225772112607956, |
| "eval_runtime": 118.3874, |
| "eval_samples_per_second": 8.81, |
| "eval_steps_per_second": 0.279, |
| "step": 68000 |
| }, |
| { |
| "epoch": 3.55, |
| "learning_rate": 1.55656668966877e-05, |
| "loss": 0.1346, |
| "step": 72000 |
| }, |
| { |
| "epoch": 3.55, |
| "eval_bleu": 42.1593, |
| "eval_gen_len": 34.5043, |
| "eval_loss": 0.10208923369646072, |
| "eval_runtime": 84.9237, |
| "eval_samples_per_second": 12.282, |
| "eval_steps_per_second": 0.389, |
| "step": 72000 |
| }, |
| { |
| "epoch": 3.75, |
| "learning_rate": 1.5319339018138805e-05, |
| "loss": 0.1351, |
| "step": 76000 |
| }, |
| { |
| "epoch": 3.75, |
| "eval_bleu": 42.0667, |
| "eval_gen_len": 35.1371, |
| "eval_loss": 0.10195324569940567, |
| "eval_runtime": 115.0559, |
| "eval_samples_per_second": 9.065, |
| "eval_steps_per_second": 0.287, |
| "step": 76000 |
| }, |
| { |
| "epoch": 3.94, |
| "learning_rate": 1.5073011139589906e-05, |
| "loss": 0.1358, |
| "step": 80000 |
| }, |
| { |
| "epoch": 3.94, |
| "eval_bleu": 42.7449, |
| "eval_gen_len": 35.2848, |
| "eval_loss": 0.10178153961896896, |
| "eval_runtime": 108.8465, |
| "eval_samples_per_second": 9.582, |
| "eval_steps_per_second": 0.303, |
| "step": 80000 |
| } |
| ], |
| "max_steps": 324608, |
| "num_train_epochs": 16, |
| "total_flos": 1.3796012702564352e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
