| { | |
| "best_metric": 44.8653, | |
| "best_model_checkpoint": "xl_ox-wn_cod_20ep_grad-acc/checkpoint-27400", | |
| "epoch": 20.0, | |
| "global_step": 27400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.75e-05, | |
| "loss": 1.8703, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_gen_len": 11.332355886139322, | |
| "eval_loss": 1.6332706212997437, | |
| "eval_rouge1": 32.6985, | |
| "eval_rouge2": 12.1019, | |
| "eval_rougeL": 30.2866, | |
| "eval_rougeLsum": 30.3052, | |
| "eval_runtime": 298.865, | |
| "eval_samples_per_second": 46.784, | |
| "eval_steps_per_second": 1.462, | |
| "step": 1370 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.5949, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_gen_len": 11.296452581891003, | |
| "eval_loss": 1.5355325937271118, | |
| "eval_rouge1": 34.2008, | |
| "eval_rouge2": 13.6349, | |
| "eval_rougeL": 31.7662, | |
| "eval_rougeLsum": 31.7895, | |
| "eval_runtime": 299.2079, | |
| "eval_samples_per_second": 46.73, | |
| "eval_steps_per_second": 1.461, | |
| "step": 2740 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.25e-05, | |
| "loss": 1.4162, | |
| "step": 4110 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_gen_len": 11.445930482048349, | |
| "eval_loss": 1.4717226028442383, | |
| "eval_rouge1": 35.9397, | |
| "eval_rouge2": 15.1721, | |
| "eval_rougeL": 33.3504, | |
| "eval_rougeLsum": 33.3788, | |
| "eval_runtime": 307.2438, | |
| "eval_samples_per_second": 45.508, | |
| "eval_steps_per_second": 1.422, | |
| "step": 4110 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4e-05, | |
| "loss": 1.2689, | |
| "step": 5480 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_gen_len": 11.853382920898298, | |
| "eval_loss": 1.4403611421585083, | |
| "eval_rouge1": 37.206, | |
| "eval_rouge2": 16.7488, | |
| "eval_rougeL": 34.5745, | |
| "eval_rougeLsum": 34.5997, | |
| "eval_runtime": 302.3654, | |
| "eval_samples_per_second": 46.242, | |
| "eval_steps_per_second": 1.445, | |
| "step": 5480 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.1461, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_gen_len": 11.862322986697182, | |
| "eval_loss": 1.43238365650177, | |
| "eval_rouge1": 38.4381, | |
| "eval_rouge2": 18.2291, | |
| "eval_rougeL": 35.7378, | |
| "eval_rougeLsum": 35.776, | |
| "eval_runtime": 312.8081, | |
| "eval_samples_per_second": 44.698, | |
| "eval_steps_per_second": 1.397, | |
| "step": 6850 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.0438, | |
| "step": 8220 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_gen_len": 11.818766986125018, | |
| "eval_loss": 1.4202848672866821, | |
| "eval_rouge1": 39.5466, | |
| "eval_rouge2": 19.5771, | |
| "eval_rougeL": 36.8922, | |
| "eval_rougeLsum": 36.9236, | |
| "eval_runtime": 315.1937, | |
| "eval_samples_per_second": 44.36, | |
| "eval_steps_per_second": 1.386, | |
| "step": 8220 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.2500000000000004e-05, | |
| "loss": 0.9541, | |
| "step": 9590 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_gen_len": 11.547775711629237, | |
| "eval_loss": 1.4285295009613037, | |
| "eval_rouge1": 40.3838, | |
| "eval_rouge2": 20.4526, | |
| "eval_rougeL": 37.7439, | |
| "eval_rougeLsum": 37.7834, | |
| "eval_runtime": 306.5558, | |
| "eval_samples_per_second": 45.61, | |
| "eval_steps_per_second": 1.426, | |
| "step": 9590 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8774, | |
| "step": 10960 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_gen_len": 11.85824631669289, | |
| "eval_loss": 1.4424927234649658, | |
| "eval_rouge1": 41.3562, | |
| "eval_rouge2": 21.7414, | |
| "eval_rougeL": 38.7297, | |
| "eval_rougeLsum": 38.7663, | |
| "eval_runtime": 300.481, | |
| "eval_samples_per_second": 46.532, | |
| "eval_steps_per_second": 1.454, | |
| "step": 10960 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 2.7500000000000004e-05, | |
| "loss": 0.8107, | |
| "step": 12330 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_gen_len": 11.772207123444428, | |
| "eval_loss": 1.4459319114685059, | |
| "eval_rouge1": 41.7443, | |
| "eval_rouge2": 22.5744, | |
| "eval_rougeL": 39.1407, | |
| "eval_rougeLsum": 39.1869, | |
| "eval_runtime": 297.6627, | |
| "eval_samples_per_second": 46.973, | |
| "eval_steps_per_second": 1.468, | |
| "step": 12330 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.7529, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_gen_len": 12.04198254899156, | |
| "eval_loss": 1.4770309925079346, | |
| "eval_rouge1": 42.415, | |
| "eval_rouge2": 23.315, | |
| "eval_rougeL": 39.8121, | |
| "eval_rougeLsum": 39.8443, | |
| "eval_runtime": 306.0301, | |
| "eval_samples_per_second": 45.688, | |
| "eval_steps_per_second": 1.428, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 2.25e-05, | |
| "loss": 0.7021, | |
| "step": 15070 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_gen_len": 12.064940637963096, | |
| "eval_loss": 1.5062141418457031, | |
| "eval_rouge1": 42.8528, | |
| "eval_rouge2": 24.038, | |
| "eval_rougeL": 40.3049, | |
| "eval_rougeLsum": 40.3369, | |
| "eval_runtime": 315.4749, | |
| "eval_samples_per_second": 44.32, | |
| "eval_steps_per_second": 1.385, | |
| "step": 15070 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.6572, | |
| "step": 16440 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_gen_len": 12.163925046488343, | |
| "eval_loss": 1.53498375415802, | |
| "eval_rouge1": 43.2449, | |
| "eval_rouge2": 24.4856, | |
| "eval_rougeL": 40.5908, | |
| "eval_rougeLsum": 40.6058, | |
| "eval_runtime": 305.4599, | |
| "eval_samples_per_second": 45.774, | |
| "eval_steps_per_second": 1.431, | |
| "step": 16440 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 1.75e-05, | |
| "loss": 0.6182, | |
| "step": 17810 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_gen_len": 12.19503647546846, | |
| "eval_loss": 1.5605404376983643, | |
| "eval_rouge1": 43.6881, | |
| "eval_rouge2": 25.1178, | |
| "eval_rougeL": 41.0796, | |
| "eval_rougeLsum": 41.1027, | |
| "eval_runtime": 313.2658, | |
| "eval_samples_per_second": 44.633, | |
| "eval_steps_per_second": 1.395, | |
| "step": 17810 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.5855, | |
| "step": 19180 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_gen_len": 12.149549420683737, | |
| "eval_loss": 1.595506191253662, | |
| "eval_rouge1": 43.9584, | |
| "eval_rouge2": 25.3084, | |
| "eval_rougeL": 41.3372, | |
| "eval_rougeLsum": 41.3654, | |
| "eval_runtime": 304.541, | |
| "eval_samples_per_second": 45.912, | |
| "eval_steps_per_second": 1.435, | |
| "step": 19180 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.5569, | |
| "step": 20550 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_gen_len": 12.273208410813904, | |
| "eval_loss": 1.6153297424316406, | |
| "eval_rouge1": 44.1726, | |
| "eval_rouge2": 25.6522, | |
| "eval_rougeL": 41.5898, | |
| "eval_rougeLsum": 41.6245, | |
| "eval_runtime": 303.9159, | |
| "eval_samples_per_second": 46.006, | |
| "eval_steps_per_second": 1.438, | |
| "step": 20550 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 1e-05, | |
| "loss": 0.5329, | |
| "step": 21920 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_gen_len": 12.136389643827778, | |
| "eval_loss": 1.642225742340088, | |
| "eval_rouge1": 44.3992, | |
| "eval_rouge2": 26.0818, | |
| "eval_rougeL": 41.8869, | |
| "eval_rougeLsum": 41.9091, | |
| "eval_runtime": 307.8415, | |
| "eval_samples_per_second": 45.419, | |
| "eval_steps_per_second": 1.42, | |
| "step": 21920 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 7.5e-06, | |
| "loss": 0.5121, | |
| "step": 23290 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_gen_len": 12.27370905449864, | |
| "eval_loss": 1.6686856746673584, | |
| "eval_rouge1": 44.6286, | |
| "eval_rouge2": 26.3925, | |
| "eval_rougeL": 42.0502, | |
| "eval_rougeLsum": 42.0707, | |
| "eval_runtime": 318.8767, | |
| "eval_samples_per_second": 43.848, | |
| "eval_steps_per_second": 1.37, | |
| "step": 23290 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 5e-06, | |
| "loss": 0.4949, | |
| "step": 24660 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_gen_len": 12.227864397081962, | |
| "eval_loss": 1.6946537494659424, | |
| "eval_rouge1": 44.6441, | |
| "eval_rouge2": 26.459, | |
| "eval_rougeL": 42.0687, | |
| "eval_rougeLsum": 42.0946, | |
| "eval_runtime": 303.6837, | |
| "eval_samples_per_second": 46.041, | |
| "eval_steps_per_second": 1.439, | |
| "step": 24660 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 2.5e-06, | |
| "loss": 0.4832, | |
| "step": 26030 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_gen_len": 12.23701902446002, | |
| "eval_loss": 1.705538034439087, | |
| "eval_rouge1": 44.775, | |
| "eval_rouge2": 26.6738, | |
| "eval_rougeL": 42.2315, | |
| "eval_rougeLsum": 42.2558, | |
| "eval_runtime": 303.9, | |
| "eval_samples_per_second": 46.009, | |
| "eval_steps_per_second": 1.438, | |
| "step": 26030 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.4758, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_gen_len": 12.328994421398942, | |
| "eval_loss": 1.7153083086013794, | |
| "eval_rouge1": 44.8653, | |
| "eval_rouge2": 26.7238, | |
| "eval_rougeL": 42.2895, | |
| "eval_rougeLsum": 42.3168, | |
| "eval_runtime": 292.7979, | |
| "eval_samples_per_second": 47.753, | |
| "eval_steps_per_second": 1.492, | |
| "step": 27400 | |
| } | |
| ], | |
| "max_steps": 27400, | |
| "num_train_epochs": 20, | |
| "total_flos": 3.652924884167688e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |