| { | |
| "best_metric": 0.8856680502646974, | |
| "best_model_checkpoint": "models_intermediate/intermediate_t5-large_25/checkpoint-44000", | |
| "epoch": 22.0, | |
| "eval_steps": 11000, | |
| "global_step": 44000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.5, | |
| "learning_rate": 3.927272727272728e-05, | |
| "loss": 0.7155, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.854545454545455e-05, | |
| "loss": 0.2765, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 3.7818181818181824e-05, | |
| "loss": 0.2224, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.709090909090909e-05, | |
| "loss": 0.1898, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.1689, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.563636363636364e-05, | |
| "loss": 0.1479, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 3.490909090909091e-05, | |
| "loss": 0.1385, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.4181818181818185e-05, | |
| "loss": 0.1274, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 3.345454545454546e-05, | |
| "loss": 0.1137, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 0.1098, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.0961, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "eval_loss": 0.13222329318523407, | |
| "eval_p-value": 0.0, | |
| "eval_runtime": 204.0215, | |
| "eval_samples_per_second": 19.606, | |
| "eval_spearman": 0.8356406711678694, | |
| "eval_steps_per_second": 2.451, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.127272727272728e-05, | |
| "loss": 0.0921, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 3.054545454545455e-05, | |
| "loss": 0.0805, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 2.9818181818181823e-05, | |
| "loss": 0.078, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 2.9090909090909093e-05, | |
| "loss": 0.0675, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 2.8363636363636366e-05, | |
| "loss": 0.0655, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 2.763636363636364e-05, | |
| "loss": 0.0572, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 2.690909090909091e-05, | |
| "loss": 0.0541, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "learning_rate": 2.6181818181818183e-05, | |
| "loss": 0.047, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 2.5454545454545457e-05, | |
| "loss": 0.048, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 10.5, | |
| "learning_rate": 2.4727272727272727e-05, | |
| "loss": 0.04, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.0377, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.16794200241565704, | |
| "eval_p-value": 0.0, | |
| "eval_runtime": 452.0989, | |
| "eval_samples_per_second": 8.848, | |
| "eval_spearman": 0.8788646157941236, | |
| "eval_steps_per_second": 1.106, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 11.5, | |
| "learning_rate": 2.3272727272727274e-05, | |
| "loss": 0.0322, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 2.2545454545454544e-05, | |
| "loss": 0.0325, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 0.0282, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 2.109090909090909e-05, | |
| "loss": 0.0276, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "learning_rate": 2.0363636363636365e-05, | |
| "loss": 0.0225, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 1.963636363636364e-05, | |
| "loss": 0.0232, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 14.5, | |
| "learning_rate": 1.8909090909090912e-05, | |
| "loss": 0.019, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.0216, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 15.5, | |
| "learning_rate": 1.7454545454545456e-05, | |
| "loss": 0.0172, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 1.672727272727273e-05, | |
| "loss": 0.0172, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 16.5, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.0137, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 16.5, | |
| "eval_loss": 0.21510018408298492, | |
| "eval_p-value": 0.0, | |
| "eval_runtime": 206.0299, | |
| "eval_samples_per_second": 19.415, | |
| "eval_spearman": 0.8823026421988669, | |
| "eval_steps_per_second": 2.427, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 1.5272727272727276e-05, | |
| "loss": 0.0144, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "learning_rate": 1.4545454545454546e-05, | |
| "loss": 0.0118, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1.381818181818182e-05, | |
| "loss": 0.0126, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 18.5, | |
| "learning_rate": 1.3090909090909092e-05, | |
| "loss": 0.0095, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 1.2363636363636364e-05, | |
| "loss": 0.0107, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 19.5, | |
| "learning_rate": 1.1636363636363637e-05, | |
| "loss": 0.0083, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 0.0089, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 20.5, | |
| "learning_rate": 1.0181818181818182e-05, | |
| "loss": 0.0076, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 9.454545454545456e-06, | |
| "loss": 0.0068, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 21.5, | |
| "learning_rate": 8.727272727272728e-06, | |
| "loss": 0.006, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.0063, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.2468361258506775, | |
| "eval_p-value": 0.0, | |
| "eval_runtime": 205.1681, | |
| "eval_samples_per_second": 19.496, | |
| "eval_spearman": 0.8856680502646974, | |
| "eval_steps_per_second": 2.437, | |
| "step": 44000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 55000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 28, | |
| "save_steps": 11000, | |
| "total_flos": 3.81040069853184e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |