| { | |
| "best_metric": 0.7993115776579469, | |
| "best_model_checkpoint": "models_intermediate/intermediate_t5-small_25/checkpoint-55000", | |
| "epoch": 27.5, | |
| "eval_steps": 11000, | |
| "global_step": 55000, | |
| "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": 1.1182, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 3.854545454545455e-05, | |
| "loss": 0.5279, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "learning_rate": 3.7818181818181824e-05, | |
| "loss": 0.4606, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3.709090909090909e-05, | |
| "loss": 0.4199, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "learning_rate": 3.6363636363636364e-05, | |
| "loss": 0.3733, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 3.563636363636364e-05, | |
| "loss": 0.3527, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 3.5, | |
| "learning_rate": 3.490909090909091e-05, | |
| "loss": 0.3254, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 3.4181818181818185e-05, | |
| "loss": 0.3012, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 3.345454545454546e-05, | |
| "loss": 0.2793, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 3.272727272727273e-05, | |
| "loss": 0.2757, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 0.2488, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 5.5, | |
| "eval_loss": 0.2022758424282074, | |
| "eval_p-value": 1.5451147277803116e-78, | |
| "eval_runtime": 127.5745, | |
| "eval_samples_per_second": 31.354, | |
| "eval_spearman": 0.2903312547350009, | |
| "eval_steps_per_second": 3.919, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 3.127272727272728e-05, | |
| "loss": 0.2514, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 6.5, | |
| "learning_rate": 3.054545454545455e-05, | |
| "loss": 0.2402, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 2.9818181818181823e-05, | |
| "loss": 0.2231, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "learning_rate": 2.9090909090909093e-05, | |
| "loss": 0.211, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 2.8363636363636366e-05, | |
| "loss": 0.2204, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 8.5, | |
| "learning_rate": 2.763636363636364e-05, | |
| "loss": 0.2083, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 2.690909090909091e-05, | |
| "loss": 0.2005, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 9.5, | |
| "learning_rate": 2.6181818181818183e-05, | |
| "loss": 0.2012, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 2.5454545454545457e-05, | |
| "loss": 0.1881, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 10.5, | |
| "learning_rate": 2.4727272727272727e-05, | |
| "loss": 0.188, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.1844, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_loss": 0.15621691942214966, | |
| "eval_p-value": 0.0, | |
| "eval_runtime": 95.3504, | |
| "eval_samples_per_second": 41.951, | |
| "eval_spearman": 0.649917736399783, | |
| "eval_steps_per_second": 5.244, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 11.5, | |
| "learning_rate": 2.3272727272727274e-05, | |
| "loss": 0.1819, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 2.2545454545454544e-05, | |
| "loss": 0.1781, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "learning_rate": 2.1818181818181818e-05, | |
| "loss": 0.174, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 2.109090909090909e-05, | |
| "loss": 0.1702, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 13.5, | |
| "learning_rate": 2.0363636363636365e-05, | |
| "loss": 0.1763, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 1.963636363636364e-05, | |
| "loss": 0.1587, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 14.5, | |
| "learning_rate": 1.8909090909090912e-05, | |
| "loss": 0.1564, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 1.8181818181818182e-05, | |
| "loss": 0.1682, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 15.5, | |
| "learning_rate": 1.7454545454545456e-05, | |
| "loss": 0.155, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 1.672727272727273e-05, | |
| "loss": 0.1603, | |
| "step": 32000 | |
| }, | |
| { | |
| "epoch": 16.5, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 0.1545, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 16.5, | |
| "eval_loss": 0.14174304902553558, | |
| "eval_p-value": 0.0, | |
| "eval_runtime": 110.3477, | |
| "eval_samples_per_second": 36.249, | |
| "eval_spearman": 0.7528888134451838, | |
| "eval_steps_per_second": 4.531, | |
| "step": 33000 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 1.5272727272727276e-05, | |
| "loss": 0.1542, | |
| "step": 34000 | |
| }, | |
| { | |
| "epoch": 17.5, | |
| "learning_rate": 1.4545454545454546e-05, | |
| "loss": 0.1479, | |
| "step": 35000 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 1.381818181818182e-05, | |
| "loss": 0.1539, | |
| "step": 36000 | |
| }, | |
| { | |
| "epoch": 18.5, | |
| "learning_rate": 1.3090909090909092e-05, | |
| "loss": 0.1495, | |
| "step": 37000 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 1.2363636363636364e-05, | |
| "loss": 0.1457, | |
| "step": 38000 | |
| }, | |
| { | |
| "epoch": 19.5, | |
| "learning_rate": 1.1636363636363637e-05, | |
| "loss": 0.1425, | |
| "step": 39000 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 1.0909090909090909e-05, | |
| "loss": 0.1494, | |
| "step": 40000 | |
| }, | |
| { | |
| "epoch": 20.5, | |
| "learning_rate": 1.0181818181818182e-05, | |
| "loss": 0.1475, | |
| "step": 41000 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 9.454545454545456e-06, | |
| "loss": 0.1375, | |
| "step": 42000 | |
| }, | |
| { | |
| "epoch": 21.5, | |
| "learning_rate": 8.727272727272728e-06, | |
| "loss": 0.1395, | |
| "step": 43000 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 0.1429, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_loss": 0.13350430130958557, | |
| "eval_p-value": 0.0, | |
| "eval_runtime": 89.7141, | |
| "eval_samples_per_second": 44.586, | |
| "eval_spearman": 0.7831815156908372, | |
| "eval_steps_per_second": 5.573, | |
| "step": 44000 | |
| }, | |
| { | |
| "epoch": 22.5, | |
| "learning_rate": 7.272727272727273e-06, | |
| "loss": 0.1416, | |
| "step": 45000 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 6.545454545454546e-06, | |
| "loss": 0.1359, | |
| "step": 46000 | |
| }, | |
| { | |
| "epoch": 23.5, | |
| "learning_rate": 5.8181818181818185e-06, | |
| "loss": 0.1419, | |
| "step": 47000 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 5.090909090909091e-06, | |
| "loss": 0.1351, | |
| "step": 48000 | |
| }, | |
| { | |
| "epoch": 24.5, | |
| "learning_rate": 4.363636363636364e-06, | |
| "loss": 0.1363, | |
| "step": 49000 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 3.6363636363636366e-06, | |
| "loss": 0.1358, | |
| "step": 50000 | |
| }, | |
| { | |
| "epoch": 25.5, | |
| "learning_rate": 2.9090909090909093e-06, | |
| "loss": 0.1381, | |
| "step": 51000 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 2.181818181818182e-06, | |
| "loss": 0.1334, | |
| "step": 52000 | |
| }, | |
| { | |
| "epoch": 26.5, | |
| "learning_rate": 1.4545454545454546e-06, | |
| "loss": 0.1367, | |
| "step": 53000 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "learning_rate": 7.272727272727273e-07, | |
| "loss": 0.1322, | |
| "step": 54000 | |
| }, | |
| { | |
| "epoch": 27.5, | |
| "learning_rate": 0.0, | |
| "loss": 0.1343, | |
| "step": 55000 | |
| }, | |
| { | |
| "epoch": 27.5, | |
| "eval_loss": 0.13135388493537903, | |
| "eval_p-value": 0.0, | |
| "eval_runtime": 88.8776, | |
| "eval_samples_per_second": 45.006, | |
| "eval_spearman": 0.7993115776579469, | |
| "eval_steps_per_second": 5.626, | |
| "step": 55000 | |
| } | |
| ], | |
| "logging_steps": 1000, | |
| "max_steps": 55000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 28, | |
| "save_steps": 11000, | |
| "total_flos": 2.9774293340258304e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |