Training logs (final)
Browse files
training_logs/per_generator_auroc.tsv
CHANGED
|
@@ -402,3 +402,133 @@ step generator AUROC TPR_at_1pct_FPR
|
|
| 402 |
150 mpt-chat 0.9813 0.7884
|
| 403 |
150 MACRO_AVG 0.9782 0.8240
|
| 404 |
150 GLOBAL_TPR_AT_1FPR - 0.8165
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 402 |
150 mpt-chat 0.9813 0.7884
|
| 403 |
150 MACRO_AVG 0.9782 0.8240
|
| 404 |
150 GLOBAL_TPR_AT_1FPR - 0.8165
|
| 405 |
+
155 chatgpt 0.9677 0.6687
|
| 406 |
+
155 cohere 0.9509 0.5434
|
| 407 |
+
155 cohere-chat 0.9707 0.4756
|
| 408 |
+
155 gpt2 0.9619 0.5992
|
| 409 |
+
155 gpt3 0.9633 0.5464
|
| 410 |
+
155 gpt4 0.9772 0.4079
|
| 411 |
+
155 llama-chat 0.9649 0.5774
|
| 412 |
+
155 mistral 0.9574 0.5872
|
| 413 |
+
155 mistral-chat 0.9628 0.6242
|
| 414 |
+
155 mpt 0.9555 0.4755
|
| 415 |
+
155 mpt-chat 0.9693 0.4884
|
| 416 |
+
155 MACRO_AVG 0.9638 0.5449
|
| 417 |
+
155 GLOBAL_TPR_AT_1FPR - 0.4704
|
| 418 |
+
160 chatgpt 0.9661 0.4106
|
| 419 |
+
160 cohere 0.9506 0.3381
|
| 420 |
+
160 cohere-chat 0.9707 0.3376
|
| 421 |
+
160 gpt2 0.9605 0.4023
|
| 422 |
+
160 gpt3 0.9618 0.2762
|
| 423 |
+
160 gpt4 0.9744 0.2229
|
| 424 |
+
160 llama-chat 0.9629 0.2906
|
| 425 |
+
160 mistral 0.9568 0.4174
|
| 426 |
+
160 mistral-chat 0.9614 0.3479
|
| 427 |
+
160 mpt 0.9551 0.3721
|
| 428 |
+
160 mpt-chat 0.9694 0.3760
|
| 429 |
+
160 MACRO_AVG 0.9627 0.3447
|
| 430 |
+
160 GLOBAL_TPR_AT_1FPR - 0.3351
|
| 431 |
+
165 chatgpt 0.9791 0.7940
|
| 432 |
+
165 cohere 0.9627 0.7253
|
| 433 |
+
165 cohere-chat 0.9802 0.7616
|
| 434 |
+
165 gpt2 0.9723 0.7623
|
| 435 |
+
165 gpt3 0.9759 0.7555
|
| 436 |
+
165 gpt4 0.9843 0.7390
|
| 437 |
+
165 llama-chat 0.9768 0.7970
|
| 438 |
+
165 mistral 0.9674 0.7208
|
| 439 |
+
165 mistral-chat 0.9743 0.7819
|
| 440 |
+
165 mpt 0.9685 0.6694
|
| 441 |
+
165 mpt-chat 0.9791 0.7321
|
| 442 |
+
165 MACRO_AVG 0.9746 0.7490
|
| 443 |
+
165 GLOBAL_TPR_AT_1FPR - 0.7259
|
| 444 |
+
170 chatgpt 0.9674 0.7011
|
| 445 |
+
170 cohere 0.9541 0.6143
|
| 446 |
+
170 cohere-chat 0.9744 0.6201
|
| 447 |
+
170 gpt2 0.9623 0.6664
|
| 448 |
+
170 gpt3 0.9623 0.6468
|
| 449 |
+
170 gpt4 0.9777 0.5003
|
| 450 |
+
170 llama-chat 0.9657 0.6875
|
| 451 |
+
170 mistral 0.9584 0.6400
|
| 452 |
+
170 mistral-chat 0.9633 0.6928
|
| 453 |
+
170 mpt 0.9569 0.5728
|
| 454 |
+
170 mpt-chat 0.9718 0.6126
|
| 455 |
+
170 MACRO_AVG 0.9649 0.6323
|
| 456 |
+
170 GLOBAL_TPR_AT_1FPR - 0.5749
|
| 457 |
+
175 chatgpt 0.9568 0.4008
|
| 458 |
+
175 cohere 0.9414 0.3177
|
| 459 |
+
175 cohere-chat 0.9663 0.3451
|
| 460 |
+
175 gpt2 0.9508 0.3894
|
| 461 |
+
175 gpt3 0.9512 0.3215
|
| 462 |
+
175 gpt4 0.9696 0.2309
|
| 463 |
+
175 llama-chat 0.9538 0.2815
|
| 464 |
+
175 mistral 0.9470 0.4098
|
| 465 |
+
175 mistral-chat 0.9527 0.3630
|
| 466 |
+
175 mpt 0.9447 0.3328
|
| 467 |
+
175 mpt-chat 0.9660 0.3603
|
| 468 |
+
175 MACRO_AVG 0.9546 0.3412
|
| 469 |
+
175 GLOBAL_TPR_AT_1FPR - 0.3331
|
| 470 |
+
180 chatgpt 0.9553 0.4242
|
| 471 |
+
180 cohere 0.9423 0.3472
|
| 472 |
+
180 cohere-chat 0.9663 0.3880
|
| 473 |
+
180 gpt2 0.9486 0.4174
|
| 474 |
+
180 gpt3 0.9505 0.3796
|
| 475 |
+
180 gpt4 0.9652 0.2288
|
| 476 |
+
180 llama-chat 0.9522 0.3109
|
| 477 |
+
180 mistral 0.9469 0.4506
|
| 478 |
+
180 mistral-chat 0.9506 0.4023
|
| 479 |
+
180 mpt 0.9435 0.3804
|
| 480 |
+
180 mpt-chat 0.9623 0.4154
|
| 481 |
+
180 MACRO_AVG 0.9531 0.3768
|
| 482 |
+
180 GLOBAL_TPR_AT_1FPR - 0.3631
|
| 483 |
+
185 chatgpt 0.9717 0.7185
|
| 484 |
+
185 cohere 0.9550 0.5842
|
| 485 |
+
185 cohere-chat 0.9762 0.7023
|
| 486 |
+
185 gpt2 0.9623 0.5887
|
| 487 |
+
185 gpt3 0.9667 0.6158
|
| 488 |
+
185 gpt4 0.9818 0.7125
|
| 489 |
+
185 llama-chat 0.9702 0.7208
|
| 490 |
+
185 mistral 0.9571 0.5728
|
| 491 |
+
185 mistral-chat 0.9675 0.6928
|
| 492 |
+
185 mpt 0.9561 0.4702
|
| 493 |
+
185 mpt-chat 0.9737 0.6415
|
| 494 |
+
185 MACRO_AVG 0.9671 0.6382
|
| 495 |
+
185 GLOBAL_TPR_AT_1FPR - 0.6511
|
| 496 |
+
190 chatgpt 0.9753 0.7713
|
| 497 |
+
190 cohere 0.9633 0.7245
|
| 498 |
+
190 cohere-chat 0.9817 0.7869
|
| 499 |
+
190 gpt2 0.9688 0.7072
|
| 500 |
+
190 gpt3 0.9701 0.7358
|
| 501 |
+
190 gpt4 0.9840 0.8045
|
| 502 |
+
190 llama-chat 0.9745 0.8121
|
| 503 |
+
190 mistral 0.9652 0.6996
|
| 504 |
+
190 mistral-chat 0.9706 0.7751
|
| 505 |
+
190 mpt 0.9650 0.6196
|
| 506 |
+
190 mpt-chat 0.9758 0.7128
|
| 507 |
+
190 MACRO_AVG 0.9722 0.7409
|
| 508 |
+
190 GLOBAL_TPR_AT_1FPR - 0.7280
|
| 509 |
+
195 chatgpt 0.9703 0.7275
|
| 510 |
+
195 cohere 0.9594 0.6498
|
| 511 |
+
195 cohere-chat 0.9783 0.6803
|
| 512 |
+
195 gpt2 0.9661 0.6702
|
| 513 |
+
195 gpt3 0.9656 0.6792
|
| 514 |
+
195 gpt4 0.9785 0.5521
|
| 515 |
+
195 llama-chat 0.9697 0.7275
|
| 516 |
+
195 mistral 0.9627 0.6453
|
| 517 |
+
195 mistral-chat 0.9659 0.7072
|
| 518 |
+
195 mpt 0.9625 0.5985
|
| 519 |
+
195 mpt-chat 0.9732 0.6254
|
| 520 |
+
195 MACRO_AVG 0.9684 0.6603
|
| 521 |
+
195 GLOBAL_TPR_AT_1FPR - 0.6130
|
| 522 |
+
200 chatgpt 0.9766 0.7638
|
| 523 |
+
200 cohere 0.9601 0.6709
|
| 524 |
+
200 cohere-chat 0.9793 0.7491
|
| 525 |
+
200 gpt2 0.9693 0.6589
|
| 526 |
+
200 gpt3 0.9716 0.6747
|
| 527 |
+
200 gpt4 0.9848 0.7539
|
| 528 |
+
200 llama-chat 0.9745 0.7879
|
| 529 |
+
200 mistral 0.9646 0.6355
|
| 530 |
+
200 mistral-chat 0.9722 0.7411
|
| 531 |
+
200 mpt 0.9656 0.5774
|
| 532 |
+
200 mpt-chat 0.9778 0.6749
|
| 533 |
+
200 MACRO_AVG 0.9724 0.6989
|
| 534 |
+
200 GLOBAL_TPR_AT_1FPR - 0.6943
|