NasimB commited on
Commit
fed2e05
·
1 Parent(s): d4eedf4

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +35 -35
README.md CHANGED
@@ -16,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the generator dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 4.4050
20
 
21
  ## Model description
22
 
@@ -49,40 +49,40 @@ The following hyperparameters were used during training:
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:-----:|:-----:|:---------------:|
52
- | 6.7153 | 0.29 | 500 | 5.6161 |
53
- | 5.3436 | 0.58 | 1000 | 5.1899 |
54
- | 4.9993 | 0.87 | 1500 | 4.9392 |
55
- | 4.7264 | 1.16 | 2000 | 4.7817 |
56
- | 4.5632 | 1.45 | 2500 | 4.6599 |
57
- | 4.4515 | 1.74 | 3000 | 4.5490 |
58
- | 4.3483 | 2.02 | 3500 | 4.4674 |
59
- | 4.1412 | 2.31 | 4000 | 4.4283 |
60
- | 4.1268 | 2.6 | 4500 | 4.3805 |
61
- | 4.0932 | 2.89 | 5000 | 4.3336 |
62
- | 3.9281 | 3.18 | 5500 | 4.3330 |
63
- | 3.8693 | 3.47 | 6000 | 4.3021 |
64
- | 3.8701 | 3.76 | 6500 | 4.2746 |
65
- | 3.8108 | 4.05 | 7000 | 4.2753 |
66
- | 3.6096 | 4.34 | 7500 | 4.2838 |
67
- | 3.6425 | 4.63 | 8000 | 4.2588 |
68
- | 3.6484 | 4.92 | 8500 | 4.2325 |
69
- | 3.4344 | 5.21 | 9000 | 4.2856 |
70
- | 3.3896 | 5.49 | 9500 | 4.2764 |
71
- | 3.4182 | 5.78 | 10000 | 4.2599 |
72
- | 3.3427 | 6.07 | 10500 | 4.2920 |
73
- | 3.1434 | 6.36 | 11000 | 4.3128 |
74
- | 3.164 | 6.65 | 11500 | 4.3048 |
75
- | 3.1778 | 6.94 | 12000 | 4.2961 |
76
- | 2.9609 | 7.23 | 12500 | 4.3472 |
77
- | 2.9349 | 7.52 | 13000 | 4.3537 |
78
- | 2.9521 | 7.81 | 13500 | 4.3518 |
79
- | 2.8837 | 8.1 | 14000 | 4.3753 |
80
- | 2.7663 | 8.39 | 14500 | 4.3885 |
81
- | 2.771 | 8.68 | 15000 | 4.3923 |
82
- | 2.7798 | 8.96 | 15500 | 4.3920 |
83
- | 2.6934 | 9.25 | 16000 | 4.4025 |
84
- | 2.685 | 9.54 | 16500 | 4.4043 |
85
- | 2.688 | 9.83 | 17000 | 4.4050 |
86
 
87
 
88
  ### Framework versions
 
16
 
17
  This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the generator dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 4.4031
20
 
21
  ## Model description
22
 
 
49
 
50
  | Training Loss | Epoch | Step | Validation Loss |
51
  |:-------------:|:-----:|:-----:|:---------------:|
52
+ | 6.7063 | 0.29 | 500 | 5.6161 |
53
+ | 5.3409 | 0.58 | 1000 | 5.1879 |
54
+ | 4.9975 | 0.87 | 1500 | 4.9292 |
55
+ | 4.7248 | 1.16 | 2000 | 4.7819 |
56
+ | 4.5625 | 1.45 | 2500 | 4.6577 |
57
+ | 4.4518 | 1.74 | 3000 | 4.5536 |
58
+ | 4.3506 | 2.02 | 3500 | 4.4718 |
59
+ | 4.1444 | 2.31 | 4000 | 4.4324 |
60
+ | 4.1299 | 2.6 | 4500 | 4.3859 |
61
+ | 4.097 | 2.89 | 5000 | 4.3383 |
62
+ | 3.9322 | 3.18 | 5500 | 4.3372 |
63
+ | 3.8738 | 3.47 | 6000 | 4.3092 |
64
+ | 3.8743 | 3.76 | 6500 | 4.2795 |
65
+ | 3.8147 | 4.05 | 7000 | 4.2758 |
66
+ | 3.6152 | 4.34 | 7500 | 4.2857 |
67
+ | 3.6479 | 4.63 | 8000 | 4.2632 |
68
+ | 3.654 | 4.92 | 8500 | 4.2380 |
69
+ | 3.4411 | 5.21 | 9000 | 4.2846 |
70
+ | 3.398 | 5.49 | 9500 | 4.2785 |
71
+ | 3.4249 | 5.78 | 10000 | 4.2628 |
72
+ | 3.3498 | 6.07 | 10500 | 4.2910 |
73
+ | 3.1525 | 6.36 | 11000 | 4.3119 |
74
+ | 3.1727 | 6.65 | 11500 | 4.3057 |
75
+ | 3.1862 | 6.94 | 12000 | 4.2985 |
76
+ | 2.9723 | 7.23 | 12500 | 4.3475 |
77
+ | 2.9448 | 7.52 | 13000 | 4.3551 |
78
+ | 2.9617 | 7.81 | 13500 | 4.3526 |
79
+ | 2.8946 | 8.1 | 14000 | 4.3748 |
80
+ | 2.7783 | 8.39 | 14500 | 4.3866 |
81
+ | 2.7819 | 8.68 | 15000 | 4.3904 |
82
+ | 2.7913 | 8.96 | 15500 | 4.3905 |
83
+ | 2.7052 | 9.25 | 16000 | 4.4009 |
84
+ | 2.6969 | 9.54 | 16500 | 4.4029 |
85
+ | 2.7 | 9.83 | 17000 | 4.4031 |
86
 
87
 
88
  ### Framework versions