Hugging Face's logo

datafreak
/
hate-phi

PEFT
Safetensors
Generated from Trainer
Model card Files
xet
Community
datafreak commited on
Commit
440a6e6
·
verified ·
1 Parent(s): a6ad6ce

HatePhi-2

Browse files
Files changed (3) hide show
  1. README.md +21 -21
  2. adapter_config.json +1 -1
  3. adapter_model.safetensors +1 -1
README.md CHANGED
@@ -16,16 +16,16 @@ should probably proofread and complete it, then remove this comment. -->
16
 
17
  This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
- - Loss: 0.2799
20
  - Classification Report: precision recall f1-score support
21
 
22
- 0 0.50 0.23 0.32 422
23
- 1 0.92 0.96 0.94 5753
24
- 2 0.84 0.85 0.85 1260
25
 
26
- accuracy 0.90 7435
27
- macro avg 0.75 0.68 0.70 7435
28
- weighted avg 0.89 0.90 0.89 7435
29
 
30
 
31
  ## Model description
@@ -60,25 +60,25 @@ The following hyperparameters were used during training:
60
 
61
  | Training Loss | Epoch | Step | Validation Loss | Classification Report |
62
  |:-------------:|:-----:|:----:|:---------------:|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|
63
- | 0.6124 | 0.37 | 25 | 0.3696 | precision recall f1-score support
64
 
65
- 0 0.44 0.03 0.05 422
66
- 1 0.90 0.95 0.93 5753
67
- 2 0.74 0.79 0.76 1260
68
 
69
- accuracy 0.87 7435
70
- macro avg 0.70 0.59 0.58 7435
71
- weighted avg 0.85 0.87 0.85 7435
72
  |
73
- | 0.3268 | 0.74 | 50 | 0.2900 | precision recall f1-score support
74
 
75
- 0 0.50 0.16 0.25 422
76
- 1 0.92 0.96 0.94 5753
77
- 2 0.82 0.85 0.84 1260
78
 
79
- accuracy 0.89 7435
80
- macro avg 0.75 0.66 0.67 7435
81
- weighted avg 0.88 0.89 0.88 7435
82
  |
83
 
84
 
 
16
 
17
  This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on an unknown dataset.
18
  It achieves the following results on the evaluation set:
19
+ - Loss: 0.3268
20
  - Classification Report: precision recall f1-score support
21
 
22
+ 0 0.57 0.08 0.14 438
23
+ 1 0.91 0.97 0.93 5755
24
+ 2 0.80 0.79 0.80 1242
25
 
26
+ accuracy 0.89 7435
27
+ macro avg 0.76 0.61 0.62 7435
28
+ weighted avg 0.87 0.89 0.87 7435
29
 
30
 
31
  ## Model description
 
60
 
61
  | Training Loss | Epoch | Step | Validation Loss | Classification Report |
62
  |:-------------:|:-----:|:----:|:---------------:|:--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------:|
63
+ | 0.8106 | 0.37 | 25 | 0.4551 | precision recall f1-score support
64
 
65
+ 0 0.18 0.03 0.04 438
66
+ 1 0.85 0.97 0.91 5755
67
+ 2 0.75 0.46 0.57 1242
68
 
69
+ accuracy 0.83 7435
70
+ macro avg 0.59 0.49 0.51 7435
71
+ weighted avg 0.79 0.83 0.80 7435
72
  |
73
+ | 0.3677 | 0.74 | 50 | 0.3374 | precision recall f1-score support
74
 
75
+ 0 0.51 0.09 0.16 438
76
+ 1 0.91 0.95 0.93 5755
77
+ 2 0.77 0.83 0.80 1242
78
 
79
+ accuracy 0.88 7435
80
+ macro avg 0.73 0.63 0.63 7435
81
+ weighted avg 0.87 0.88 0.87 7435
82
  |
83
 
84
 
adapter_config.json CHANGED
@@ -24,8 +24,8 @@
24
  "revision": null,
25
  "target_modules": [
26
  "q_proj",
27
- "k_proj",
28
  "v_proj",
 
29
  "dense"
30
  ],
31
  "task_type": "SEQ_CLS",
 
24
  "revision": null,
25
  "target_modules": [
26
  "q_proj",
 
27
  "v_proj",
28
+ "k_proj",
29
  "dense"
30
  ],
31
  "task_type": "SEQ_CLS",
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ccf6c03f8d38720ddd510193289fda367ae521b296d349a80ef51bd8c70dbec1
3
  size 2685728
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:da022dcbd1489e36dc0f1bf17c2c399ce3df742c6bfb821af295f54bd39e7d14
3
  size 2685728