| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 13.636363636363637, | |
| "eval_steps": 200, | |
| "global_step": 600, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.022727272727272728, | |
| "eval_loss": 2.4719977378845215, | |
| "eval_runtime": 13.2838, | |
| "eval_samples_per_second": 112.995, | |
| "eval_steps_per_second": 5.721, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.22727272727272727, | |
| "grad_norm": 6.96875, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 3.6559, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.45454545454545453, | |
| "grad_norm": 3.734375, | |
| "learning_rate": 0.00013333333333333334, | |
| "loss": 2.7443, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6818181818181818, | |
| "grad_norm": 2.890625, | |
| "learning_rate": 0.0002, | |
| "loss": 2.6549, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 2.484375, | |
| "learning_rate": 0.00019984815164333163, | |
| "loss": 2.7133, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.1363636363636362, | |
| "grad_norm": 3.140625, | |
| "learning_rate": 0.00019939306773179497, | |
| "loss": 2.3048, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.3636363636363638, | |
| "grad_norm": 2.53125, | |
| "learning_rate": 0.00019863613034027224, | |
| "loss": 2.0044, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.5909090909090908, | |
| "grad_norm": 2.15625, | |
| "learning_rate": 0.00019757963826274357, | |
| "loss": 2.0418, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 1.75, | |
| "learning_rate": 0.00019622680003092503, | |
| "loss": 2.0792, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.0454545454545454, | |
| "grad_norm": 2.40625, | |
| "learning_rate": 0.00019458172417006347, | |
| "loss": 1.9592, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.2727272727272725, | |
| "grad_norm": 1.9609375, | |
| "learning_rate": 0.00019264940672148018, | |
| "loss": 1.206, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 2.078125, | |
| "learning_rate": 0.00019043571606975777, | |
| "loss": 1.2323, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.7272727272727275, | |
| "grad_norm": 1.7578125, | |
| "learning_rate": 0.0001879473751206489, | |
| "loss": 1.2829, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.9545454545454546, | |
| "grad_norm": 1.6953125, | |
| "learning_rate": 0.00018519194088383273, | |
| "loss": 1.3162, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.1818181818181817, | |
| "grad_norm": 1.8984375, | |
| "learning_rate": 0.0001821777815225245, | |
| "loss": 0.7771, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.409090909090909, | |
| "grad_norm": 1.5390625, | |
| "learning_rate": 0.00017891405093963938, | |
| "loss": 0.643, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 1.8046875, | |
| "learning_rate": 0.00017541066097768963, | |
| "loss": 0.6739, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 3.8636363636363638, | |
| "grad_norm": 1.5234375, | |
| "learning_rate": 0.00017167825131684513, | |
| "loss": 0.709, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.090909090909091, | |
| "grad_norm": 1.234375, | |
| "learning_rate": 0.00016772815716257412, | |
| "loss": 0.567, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.318181818181818, | |
| "grad_norm": 1.265625, | |
| "learning_rate": 0.00016357237482099684, | |
| "loss": 0.3307, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "grad_norm": 1.1875, | |
| "learning_rate": 0.00015922352526649803, | |
| "loss": 0.3485, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.545454545454545, | |
| "eval_loss": 3.4570090770721436, | |
| "eval_runtime": 13.1001, | |
| "eval_samples_per_second": 114.579, | |
| "eval_steps_per_second": 5.801, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 4.7727272727272725, | |
| "grad_norm": 1.21875, | |
| "learning_rate": 0.00015469481581224272, | |
| "loss": 0.3798, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 1.40625, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.3946, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.2272727272727275, | |
| "grad_norm": 0.92578125, | |
| "learning_rate": 0.00014515333583108896, | |
| "loss": 0.185, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.454545454545454, | |
| "grad_norm": 0.953125, | |
| "learning_rate": 0.00014016954246529696, | |
| "loss": 0.2086, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 5.681818181818182, | |
| "grad_norm": 0.9609375, | |
| "learning_rate": 0.00013506375551927547, | |
| "loss": 0.2037, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 5.909090909090909, | |
| "grad_norm": 0.94921875, | |
| "learning_rate": 0.00012985148110016947, | |
| "loss": 0.2147, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.136363636363637, | |
| "grad_norm": 0.67578125, | |
| "learning_rate": 0.00012454854871407994, | |
| "loss": 0.1516, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 6.363636363636363, | |
| "grad_norm": 0.64453125, | |
| "learning_rate": 0.00011917106319237386, | |
| "loss": 0.1077, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 6.590909090909091, | |
| "grad_norm": 0.640625, | |
| "learning_rate": 0.00011373535578184082, | |
| "loss": 0.1024, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 6.818181818181818, | |
| "grad_norm": 0.62109375, | |
| "learning_rate": 0.00010825793454723325, | |
| "loss": 0.0919, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.045454545454546, | |
| "grad_norm": 0.392578125, | |
| "learning_rate": 0.00010275543423681621, | |
| "loss": 0.075, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 7.2727272727272725, | |
| "grad_norm": 0.3515625, | |
| "learning_rate": 9.724456576318381e-05, | |
| "loss": 0.0312, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 0.3203125, | |
| "learning_rate": 9.174206545276677e-05, | |
| "loss": 0.0276, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 7.7272727272727275, | |
| "grad_norm": 0.333984375, | |
| "learning_rate": 8.626464421815919e-05, | |
| "loss": 0.0259, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 7.954545454545455, | |
| "grad_norm": 0.30859375, | |
| "learning_rate": 8.082893680762619e-05, | |
| "loss": 0.0246, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 8.181818181818182, | |
| "grad_norm": 0.22265625, | |
| "learning_rate": 7.54514512859201e-05, | |
| "loss": 0.0135, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 8.409090909090908, | |
| "grad_norm": 0.2294921875, | |
| "learning_rate": 7.014851889983057e-05, | |
| "loss": 0.0107, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 8.636363636363637, | |
| "grad_norm": 0.1884765625, | |
| "learning_rate": 6.493624448072457e-05, | |
| "loss": 0.0105, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 8.863636363636363, | |
| "grad_norm": 0.267578125, | |
| "learning_rate": 5.983045753470308e-05, | |
| "loss": 0.0098, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "grad_norm": 0.14453125, | |
| "learning_rate": 5.484666416891109e-05, | |
| "loss": 0.009, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.090909090909092, | |
| "eval_loss": 4.0009965896606445, | |
| "eval_runtime": 13.0419, | |
| "eval_samples_per_second": 115.091, | |
| "eval_steps_per_second": 5.827, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 9.318181818181818, | |
| "grad_norm": 0.171875, | |
| "learning_rate": 5.000000000000002e-05, | |
| "loss": 0.0071, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 9.545454545454545, | |
| "grad_norm": 0.16796875, | |
| "learning_rate": 4.530518418775733e-05, | |
| "loss": 0.0072, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 9.772727272727273, | |
| "grad_norm": 0.2392578125, | |
| "learning_rate": 4.077647473350201e-05, | |
| "loss": 0.0071, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 0.1572265625, | |
| "learning_rate": 3.642762517900322e-05, | |
| "loss": 0.0069, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 10.227272727272727, | |
| "grad_norm": 0.1904296875, | |
| "learning_rate": 3.227184283742591e-05, | |
| "loss": 0.0066, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 10.454545454545455, | |
| "grad_norm": 0.1416015625, | |
| "learning_rate": 2.8321748683154893e-05, | |
| "loss": 0.0066, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 10.681818181818182, | |
| "grad_norm": 0.15625, | |
| "learning_rate": 2.4589339022310386e-05, | |
| "loss": 0.0067, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 10.909090909090908, | |
| "grad_norm": 0.1376953125, | |
| "learning_rate": 2.1085949060360654e-05, | |
| "loss": 0.0062, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 11.136363636363637, | |
| "grad_norm": 0.15625, | |
| "learning_rate": 1.7822218477475494e-05, | |
| "loss": 0.0065, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 11.363636363636363, | |
| "grad_norm": 0.130859375, | |
| "learning_rate": 1.4808059116167305e-05, | |
| "loss": 0.0062, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 11.590909090909092, | |
| "grad_norm": 0.134765625, | |
| "learning_rate": 1.2052624879351104e-05, | |
| "loss": 0.0061, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 11.818181818181818, | |
| "grad_norm": 0.1455078125, | |
| "learning_rate": 9.564283930242257e-06, | |
| "loss": 0.0064, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 12.045454545454545, | |
| "grad_norm": 0.12255859375, | |
| "learning_rate": 7.350593278519824e-06, | |
| "loss": 0.0059, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 12.272727272727273, | |
| "grad_norm": 0.162109375, | |
| "learning_rate": 5.418275829936537e-06, | |
| "loss": 0.006, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 12.5, | |
| "grad_norm": 0.1318359375, | |
| "learning_rate": 3.7731999690749585e-06, | |
| "loss": 0.0061, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 12.727272727272727, | |
| "grad_norm": 0.11962890625, | |
| "learning_rate": 2.420361737256438e-06, | |
| "loss": 0.0064, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 12.954545454545455, | |
| "grad_norm": 0.1376953125, | |
| "learning_rate": 1.3638696597277679e-06, | |
| "loss": 0.0061, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 13.181818181818182, | |
| "grad_norm": 0.158203125, | |
| "learning_rate": 6.069322682050516e-07, | |
| "loss": 0.0063, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 13.409090909090908, | |
| "grad_norm": 0.11767578125, | |
| "learning_rate": 1.518483566683826e-07, | |
| "loss": 0.006, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 13.636363636363637, | |
| "grad_norm": 0.15625, | |
| "learning_rate": 0.0, | |
| "loss": 0.0063, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 13.636363636363637, | |
| "eval_loss": 4.043154239654541, | |
| "eval_runtime": 13.032, | |
| "eval_samples_per_second": 115.178, | |
| "eval_steps_per_second": 5.832, | |
| "step": 600 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 600, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 14, | |
| "save_steps": 200, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 4.240152158483251e+17, | |
| "train_batch_size": 5, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |