Hugging Face's logo

tthhanh
/
legal_llama

PEFT
Safetensors
Generated from Trainer
Model card Files
xet
Community
tthhanh commited on
Commit
31590a0
·
verified ·
1 Parent(s): dc1a4a8

End of training

Browse files
Files changed (2) hide show
  1. README.md +88 -3
  2. adapter_model.safetensors +1 -1
README.md CHANGED
@@ -1,3 +1,88 @@
1
- ---
2
- license: cc-by-3.0
3
- ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: meta-llama/Llama-2-7b-hf
3
+ library_name: peft
4
+ license: llama2
5
+ tags:
6
+ - generated_from_trainer
7
+ model-index:
8
+ - name: legal_llama
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ # legal_llama
16
+
17
+ This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on an unknown dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 0.4084
20
+ - Law Precision: 0.3274
21
+ - Law Recall: 0.5
22
+ - Law F1: 0.3957
23
+ - Law Number: 74
24
+ - Violated by Precision: 0.2857
25
+ - Violated by Recall: 0.5352
26
+ - Violated by F1: 0.3725
27
+ - Violated by Number: 71
28
+ - Violated on Precision: 0.1014
29
+ - Violated on Recall: 0.14
30
+ - Violated on F1: 0.1176
31
+ - Violated on Number: 50
32
+ - Violation Precision: 0.1545
33
+ - Violation Recall: 0.3049
34
+ - Violation F1: 0.2051
35
+ - Violation Number: 597
36
+ - Overall Precision: 0.1768
37
+ - Overall Recall: 0.3333
38
+ - Overall F1: 0.2311
39
+ - Overall Accuracy: 0.8885
40
+
41
+ ## Model description
42
+
43
+ More information needed
44
+
45
+ ## Intended uses & limitations
46
+
47
+ More information needed
48
+
49
+ ## Training and evaluation data
50
+
51
+ More information needed
52
+
53
+ ## Training procedure
54
+
55
+ ### Training hyperparameters
56
+
57
+ The following hyperparameters were used during training:
58
+ - learning_rate: 0.0001
59
+ - train_batch_size: 16
60
+ - eval_batch_size: 16
61
+ - seed: 42
62
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
63
+ - lr_scheduler_type: linear
64
+ - num_epochs: 10
65
+
66
+ ### Training results
67
+
68
+ | Training Loss | Epoch | Step | Validation Loss | Law Precision | Law Recall | Law F1 | Law Number | Violated by Precision | Violated by Recall | Violated by F1 | Violated by Number | Violated on Precision | Violated on Recall | Violated on F1 | Violated on Number | Violation Precision | Violation Recall | Violation F1 | Violation Number | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy |
69
+ |:-------------:|:-----:|:----:|:---------------:|:-------------:|:----------:|:------:|:----------:|:---------------------:|:------------------:|:--------------:|:------------------:|:---------------------:|:------------------:|:--------------:|:------------------:|:-------------------:|:----------------:|:------------:|:----------------:|:-----------------:|:--------------:|:----------:|:----------------:|
70
+ | No log | 1.0 | 45 | 0.6199 | 0.0 | 0.0 | 0.0 | 74 | 0.0 | 0.0 | 0.0 | 71 | 0.0 | 0.0 | 0.0 | 50 | 0.0023 | 0.0034 | 0.0027 | 597 | 0.0023 | 0.0025 | 0.0024 | 0.7695 |
71
+ | No log | 2.0 | 90 | 0.6097 | 0.0211 | 0.0270 | 0.0237 | 74 | 0.0 | 0.0 | 0.0 | 71 | 0.0 | 0.0 | 0.0 | 50 | 0.0171 | 0.0251 | 0.0203 | 597 | 0.0174 | 0.0215 | 0.0192 | 0.7989 |
72
+ | No log | 3.0 | 135 | 0.4495 | 0.0317 | 0.0270 | 0.0292 | 74 | 0.0 | 0.0 | 0.0 | 71 | 0.0 | 0.0 | 0.0 | 50 | 0.0426 | 0.0888 | 0.0576 | 597 | 0.0418 | 0.0694 | 0.0522 | 0.8421 |
73
+ | No log | 4.0 | 180 | 0.4742 | 0.2245 | 0.1486 | 0.1789 | 74 | 0.25 | 0.0704 | 0.1099 | 71 | 0.0 | 0.0 | 0.0 | 50 | 0.0309 | 0.0536 | 0.0392 | 597 | 0.0434 | 0.0606 | 0.0506 | 0.8416 |
74
+ | No log | 5.0 | 225 | 0.3946 | 0.1714 | 0.1622 | 0.1667 | 74 | 0.3968 | 0.3521 | 0.3731 | 71 | 0.0 | 0.0 | 0.0 | 50 | 0.1422 | 0.3317 | 0.1991 | 597 | 0.1528 | 0.2967 | 0.2017 | 0.8663 |
75
+ | No log | 6.0 | 270 | 0.3872 | 0.2278 | 0.2432 | 0.2353 | 74 | 0.2857 | 0.3099 | 0.2973 | 71 | 0.2143 | 0.06 | 0.0938 | 50 | 0.1115 | 0.2395 | 0.1521 | 597 | 0.1280 | 0.2348 | 0.1657 | 0.8742 |
76
+ | No log | 7.0 | 315 | 0.3722 | 0.2936 | 0.4324 | 0.3497 | 74 | 0.3065 | 0.5352 | 0.3897 | 71 | 0.0980 | 0.1 | 0.0990 | 50 | 0.1255 | 0.2412 | 0.1651 | 597 | 0.1530 | 0.2765 | 0.1970 | 0.8848 |
77
+ | No log | 8.0 | 360 | 0.4131 | 0.2917 | 0.4730 | 0.3608 | 74 | 0.2835 | 0.5070 | 0.3636 | 71 | 0.1017 | 0.12 | 0.1101 | 50 | 0.1329 | 0.2513 | 0.1738 | 597 | 0.1582 | 0.2866 | 0.2039 | 0.8812 |
78
+ | No log | 9.0 | 405 | 0.3990 | 0.3008 | 0.5 | 0.3756 | 74 | 0.2529 | 0.6197 | 0.3592 | 71 | 0.0909 | 0.14 | 0.1102 | 50 | 0.1425 | 0.2982 | 0.1928 | 597 | 0.1639 | 0.3359 | 0.2203 | 0.8864 |
79
+ | No log | 10.0 | 450 | 0.4084 | 0.3274 | 0.5 | 0.3957 | 74 | 0.2857 | 0.5352 | 0.3725 | 71 | 0.1014 | 0.14 | 0.1176 | 50 | 0.1545 | 0.3049 | 0.2051 | 597 | 0.1768 | 0.3333 | 0.2311 | 0.8885 |
80
+
81
+
82
+ ### Framework versions
83
+
84
+ - PEFT 0.12.0
85
+ - Transformers 4.44.0
86
+ - Pytorch 2.4.0+cu121
87
+ - Datasets 2.20.0
88
+ - Tokenizers 0.19.1
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5d77857ed8f391930673d9072b4e1fbdc9f483db677e38ae89ffa16d9ecaa6eb
3
  size 67200050
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:74eaca9d374dc627fee9d719a2059a2a737232c4a16bfd9a64943d08abe2ab40
3
  size 67200050