xvisox commited on
Commit
2df8f5d
·
verified ·
1 Parent(s): 2b03761

End of training

Browse files
Files changed (1) hide show
  1. README.md +42 -42
README.md CHANGED
@@ -14,7 +14,7 @@ should probably proofread and complete it, then remove this comment. -->
14
 
15
  This model is a fine-tuned version of [](https://huggingface.co/) on the None dataset.
16
  It achieves the following results on the evaluation set:
17
- - Loss: 0.6172
18
 
19
  ## Model description
20
 
@@ -45,51 +45,51 @@ The following hyperparameters were used during training:
45
 
46
  | Training Loss | Epoch | Step | Validation Loss |
47
  |:-------------:|:-----:|:----:|:---------------:|
48
- | 3.4882 | 1.0 | 5 | 2.8861 |
49
- | 2.5629 | 2.0 | 10 | 2.1270 |
50
- | 1.9292 | 3.0 | 15 | 1.7169 |
51
- | 1.6672 | 4.0 | 20 | 1.6025 |
52
- | 1.5733 | 5.0 | 25 | 1.5584 |
53
- | 1.5482 | 6.0 | 30 | 1.6137 |
54
- | 1.5415 | 7.0 | 35 | 1.5238 |
55
- | 1.4927 | 8.0 | 40 | 1.4616 |
56
- | 1.4387 | 9.0 | 45 | 1.4843 |
57
- | 1.4567 | 10.0 | 50 | 1.4316 |
58
- | 1.4003 | 11.0 | 55 | 1.3734 |
59
- | 1.3508 | 12.0 | 60 | 1.3262 |
60
- | 1.3077 | 13.0 | 65 | 1.2710 |
61
- | 1.3015 | 14.0 | 70 | 1.3736 |
62
- | 1.2897 | 15.0 | 75 | 1.2047 |
63
- | 1.2017 | 16.0 | 80 | 1.1661 |
64
- | 1.1367 | 17.0 | 85 | 1.1048 |
65
- | 1.1049 | 18.0 | 90 | 1.0427 |
66
- | 1.0524 | 19.0 | 95 | 0.9892 |
67
- | 1.0018 | 20.0 | 100 | 0.9464 |
68
- | 0.9639 | 21.0 | 105 | 0.9013 |
69
- | 0.9444 | 22.0 | 110 | 0.8804 |
70
- | 0.9066 | 23.0 | 115 | 0.8519 |
71
- | 0.8921 | 24.0 | 120 | 0.8233 |
72
- | 0.8562 | 25.0 | 125 | 0.8070 |
73
- | 0.8385 | 26.0 | 130 | 0.7824 |
74
- | 0.8184 | 27.0 | 135 | 0.7700 |
75
- | 0.8051 | 28.0 | 140 | 0.7532 |
76
- | 0.7883 | 29.0 | 145 | 0.7297 |
77
- | 0.7673 | 30.0 | 150 | 0.7146 |
78
- | 0.7501 | 31.0 | 155 | 0.6967 |
79
- | 0.7375 | 32.0 | 160 | 0.6826 |
80
- | 0.7204 | 33.0 | 165 | 0.6654 |
81
- | 0.7074 | 34.0 | 170 | 0.6530 |
82
- | 0.6984 | 35.0 | 175 | 0.6476 |
83
- | 0.6935 | 36.0 | 180 | 0.6368 |
84
- | 0.6802 | 37.0 | 185 | 0.6299 |
85
- | 0.6760 | 38.0 | 190 | 0.6226 |
86
- | 0.6720 | 39.0 | 195 | 0.6198 |
87
- | 0.6676 | 40.0 | 200 | 0.6172 |
88
 
89
 
90
  ### Framework versions
91
 
92
  - Transformers 5.0.0
93
- - Pytorch 2.10.0+cpu
94
  - Datasets 4.0.0
95
  - Tokenizers 0.22.2
 
14
 
15
  This model is a fine-tuned version of [](https://huggingface.co/) on the None dataset.
16
  It achieves the following results on the evaluation set:
17
+ - Loss: 0.1433
18
 
19
  ## Model description
20
 
 
45
 
46
  | Training Loss | Epoch | Step | Validation Loss |
47
  |:-------------:|:-----:|:----:|:---------------:|
48
+ | 2.9812 | 1.0 | 6 | 2.2407 |
49
+ | 2.0270 | 2.0 | 12 | 1.7466 |
50
+ | 1.5714 | 3.0 | 18 | 1.2953 |
51
+ | 1.2001 | 4.0 | 24 | 1.0696 |
52
+ | 1.0133 | 5.0 | 30 | 0.9091 |
53
+ | 0.8661 | 6.0 | 36 | 0.7848 |
54
+ | 0.7555 | 7.0 | 42 | 0.6814 |
55
+ | 0.6796 | 8.0 | 48 | 0.6318 |
56
+ | 0.6473 | 9.0 | 54 | 0.6153 |
57
+ | 0.6041 | 10.0 | 60 | 0.5501 |
58
+ | 0.5619 | 11.0 | 66 | 0.5469 |
59
+ | 0.5457 | 12.0 | 72 | 0.5018 |
60
+ | 0.5004 | 13.0 | 78 | 0.4598 |
61
+ | 0.4727 | 14.0 | 84 | 0.4299 |
62
+ | 0.4530 | 15.0 | 90 | 0.4329 |
63
+ | 0.4326 | 16.0 | 96 | 0.4042 |
64
+ | 0.4063 | 17.0 | 102 | 0.3745 |
65
+ | 0.3781 | 18.0 | 108 | 0.3800 |
66
+ | 0.3919 | 19.0 | 114 | 0.3520 |
67
+ | 0.3600 | 20.0 | 120 | 0.3237 |
68
+ | 0.3449 | 21.0 | 126 | 0.2963 |
69
+ | 0.3199 | 22.0 | 132 | 0.3008 |
70
+ | 0.3220 | 23.0 | 138 | 0.2882 |
71
+ | 0.3037 | 24.0 | 144 | 0.2534 |
72
+ | 0.2746 | 25.0 | 150 | 0.2573 |
73
+ | 0.2700 | 26.0 | 156 | 0.2359 |
74
+ | 0.2573 | 27.0 | 162 | 0.2204 |
75
+ | 0.2392 | 28.0 | 168 | 0.2122 |
76
+ | 0.2339 | 29.0 | 174 | 0.2000 |
77
+ | 0.2208 | 30.0 | 180 | 0.1913 |
78
+ | 0.2159 | 31.0 | 186 | 0.1816 |
79
+ | 0.1982 | 32.0 | 192 | 0.1747 |
80
+ | 0.1967 | 33.0 | 198 | 0.1665 |
81
+ | 0.1868 | 34.0 | 204 | 0.1642 |
82
+ | 0.1804 | 35.0 | 210 | 0.1589 |
83
+ | 0.1797 | 36.0 | 216 | 0.1555 |
84
+ | 0.1731 | 37.0 | 222 | 0.1524 |
85
+ | 0.1698 | 38.0 | 228 | 0.1471 |
86
+ | 0.1679 | 39.0 | 234 | 0.1440 |
87
+ | 0.1667 | 40.0 | 240 | 0.1433 |
88
 
89
 
90
  ### Framework versions
91
 
92
  - Transformers 5.0.0
93
+ - Pytorch 2.10.0+cu128
94
  - Datasets 4.0.0
95
  - Tokenizers 0.22.2