Hugging Face's logo

Farjfar
/
llama-2-ner

PEFT
Safetensors
Generated from Trainer
Model card Files Community
Farjfar commited on
Commit
6597297
•
1 Parent(s): 3c2d753

End of training

Browse files
Files changed (2) hide show
  1. README.md +21 -16
  2. adapter_model.safetensors +1 -1
README.md CHANGED
@@ -20,11 +20,11 @@ should probably proofread and complete it, then remove this comment. -->
20
 
21
  This model is a fine-tuned version of [NousResearch/Llama-2-7b-hf](https://huggingface.co/NousResearch/Llama-2-7b-hf) on an unknown dataset.
22
  It achieves the following results on the evaluation set:
23
- - Loss: 0.1511
24
- - Precision: 0.4375
25
- - Recall: 0.4789
26
- - F1: 0.4573
27
- - Accuracy: 0.9752
28
 
29
  ## Model description
30
 
@@ -49,22 +49,27 @@ The following hyperparameters were used during training:
49
  - seed: 42
50
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
51
  - lr_scheduler_type: linear
52
- - num_epochs: 10
53
 
54
  ### Training results
55
 
56
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
57
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
58
- | No log | 1.0 | 39 | 0.1580 | 0.0 | 0.0 | 0.0 | 0.9677 |
59
- | No log | 2.0 | 78 | 0.1313 | 0.2037 | 0.1158 | 0.1477 | 0.9654 |
60
- | No log | 3.0 | 117 | 0.0948 | 0.3494 | 0.1526 | 0.2125 | 0.9698 |
61
- | No log | 4.0 | 156 | 0.0952 | 0.2252 | 0.3947 | 0.2868 | 0.9676 |
62
- | No log | 5.0 | 195 | 0.0850 | 0.3160 | 0.3526 | 0.3333 | 0.9733 |
63
- | No log | 6.0 | 234 | 0.1012 | 0.2990 | 0.4895 | 0.3713 | 0.9690 |
64
- | No log | 7.0 | 273 | 0.1282 | 0.4476 | 0.4947 | 0.4700 | 0.9739 |
65
- | No log | 8.0 | 312 | 0.1400 | 0.4332 | 0.4947 | 0.4619 | 0.9748 |
66
- | No log | 9.0 | 351 | 0.1492 | 0.4279 | 0.5 | 0.4612 | 0.9753 |
67
- | No log | 10.0 | 390 | 0.1511 | 0.4375 | 0.4789 | 0.4573 | 0.9752 |
 
 
 
 
 
68
 
69
 
70
  ### Framework versions
 
20
 
21
  This model is a fine-tuned version of [NousResearch/Llama-2-7b-hf](https://huggingface.co/NousResearch/Llama-2-7b-hf) on an unknown dataset.
22
  It achieves the following results on the evaluation set:
23
+ - Loss: 0.1763
24
+ - Precision: 0.5149
25
+ - Recall: 0.5474
26
+ - F1: 0.5306
27
+ - Accuracy: 0.9771
28
 
29
  ## Model description
30
 
 
49
  - seed: 42
50
  - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
51
  - lr_scheduler_type: linear
52
+ - num_epochs: 15
53
 
54
  ### Training results
55
 
56
  | Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
57
  |:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:|
58
+ | No log | 1.0 | 39 | 0.1373 | 0.3704 | 0.1053 | 0.1639 | 0.9658 |
59
+ | No log | 2.0 | 78 | 0.1236 | 0.2123 | 0.3632 | 0.2680 | 0.9638 |
60
+ | No log | 3.0 | 117 | 0.0959 | 0.5333 | 0.1263 | 0.2043 | 0.9701 |
61
+ | No log | 4.0 | 156 | 0.0803 | 0.3480 | 0.5 | 0.4104 | 0.9715 |
62
+ | No log | 5.0 | 195 | 0.1118 | 0.4684 | 0.3895 | 0.4253 | 0.9758 |
63
+ | No log | 6.0 | 234 | 0.0944 | 0.4421 | 0.4421 | 0.4421 | 0.9752 |
64
+ | No log | 7.0 | 273 | 0.1037 | 0.4976 | 0.5368 | 0.5165 | 0.9772 |
65
+ | No log | 8.0 | 312 | 0.1405 | 0.4798 | 0.5632 | 0.5182 | 0.9770 |
66
+ | No log | 9.0 | 351 | 0.1623 | 0.52 | 0.5474 | 0.5333 | 0.9766 |
67
+ | No log | 10.0 | 390 | 0.1656 | 0.4906 | 0.5474 | 0.5174 | 0.9766 |
68
+ | No log | 11.0 | 429 | 0.1729 | 0.5098 | 0.5474 | 0.5279 | 0.9770 |
69
+ | No log | 12.0 | 468 | 0.1751 | 0.5073 | 0.5474 | 0.5266 | 0.9770 |
70
+ | 0.0487 | 13.0 | 507 | 0.1758 | 0.5099 | 0.5421 | 0.5255 | 0.9770 |
71
+ | 0.0487 | 14.0 | 546 | 0.1762 | 0.5073 | 0.5474 | 0.5266 | 0.9770 |
72
+ | 0.0487 | 15.0 | 585 | 0.1763 | 0.5149 | 0.5474 | 0.5306 | 0.9771 |
73
 
74
 
75
  ### Framework versions
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d1287c7076697231b1398628df19c171f6c5911bc1b8a08e19fb968b61921fb4
3
  size 33596518
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4a9c90e124e84371a7592bdfad1b9cd2a0779fda34f57796f239385d8c9df3b9
3
  size 33596518