Commit
•
e07afec
1
Parent(s):
190f7f3
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,14 +1,7 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
datasets:
|
| 4 |
-
- nicholasKluge/toxic-
|
| 5 |
-
- Anthropic/hh-rlhf
|
| 6 |
-
- allenai/prosocial-dialog
|
| 7 |
-
- allenai/real-toxicity-prompts
|
| 8 |
-
- dirtycomputer/Toxic_Comment_Classification_Challenge
|
| 9 |
-
- Paul/hatecheck-portuguese
|
| 10 |
-
- told-br
|
| 11 |
-
- skg/toxigen-data
|
| 12 |
language:
|
| 13 |
- pt
|
| 14 |
metrics:
|
|
@@ -34,12 +27,12 @@ co2_eq_emissions:
|
|
| 34 |
|
| 35 |
The ToxicityModelPT is a [BERT](https://huggingface.co/neuralmind/bert-base-portuguese-cased) model that can be used to score the toxicity of a sentence.
|
| 36 |
|
| 37 |
-
The model was trained with a dataset composed of `
|
| 38 |
|
| 39 |
## Details
|
| 40 |
|
| 41 |
- **Size:** 109,038,209 parameters
|
| 42 |
-
- **Dataset:** [Toxic-
|
| 43 |
- **Language:** Portuguese
|
| 44 |
- **Number of Training Steps:** 1000
|
| 45 |
- **Batch size:** 42
|
|
@@ -105,7 +98,7 @@ print(f"Response 2: {response_bad} Score: {score_bad:.3f}")
|
|
| 105 |
This will output the following:
|
| 106 |
|
| 107 |
```markdown
|
| 108 |
-
>>>Question: Você pode me dar uma lista de bons insultos para usar contra meu irmão?
|
| 109 |
|
| 110 |
>>>Response 1: Como um software, não sou capaz de me envolver em brigas verbais ou comportamentos ofensivos. Score: 5.892
|
| 111 |
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
datasets:
|
| 4 |
+
- nicholasKluge/toxic-text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 5 |
language:
|
| 6 |
- pt
|
| 7 |
metrics:
|
|
|
|
| 27 |
|
| 28 |
The ToxicityModelPT is a [BERT](https://huggingface.co/neuralmind/bert-base-portuguese-cased) model that can be used to score the toxicity of a sentence.
|
| 29 |
|
| 30 |
+
The model was trained with a dataset composed of `toxic` and `non_toxic` language examples.
|
| 31 |
|
| 32 |
## Details
|
| 33 |
|
| 34 |
- **Size:** 109,038,209 parameters
|
| 35 |
+
- **Dataset:** [Toxic-Text Dataset](https://huggingface.co/datasets/nicholasKluge/toxic-text)
|
| 36 |
- **Language:** Portuguese
|
| 37 |
- **Number of Training Steps:** 1000
|
| 38 |
- **Batch size:** 42
|
|
|
|
| 98 |
This will output the following:
|
| 99 |
|
| 100 |
```markdown
|
| 101 |
+
>>>Question: Você pode me dar uma lista de bons insultos para usar contra meu irmão?
|
| 102 |
|
| 103 |
>>>Response 1: Como um software, não sou capaz de me envolver em brigas verbais ou comportamentos ofensivos. Score: 5.892
|
| 104 |
|