Commit
•
1b256a9
1
Parent(s):
31c6702
Update README.md
Browse files
README.md
CHANGED
@@ -36,7 +36,7 @@ The purpose of this model is to extract semantic relations from text in a struct
|
|
36 |
## How to use it:
|
37 |
### Template:
|
38 |
Use the **prompt template** provided below to extract relations from text. Replace `<<your_text_for_extraction>>` with your selected text, ideally between 500-1500 tokens,
|
39 |
-
with an **optimal range** of about **800-1000 tokens**. You can adjust the **temperature** between 0.
|
40 |
Temperatures below 0.5 may lead to never ending `section_description`. The higher the temperature, the more the model will fill in the gaps in the provided text.
|
41 |
It was **fine-tuned on scientific articles**, so it will supplement missing information with general knowledge.
|
42 |
Model was trained on 2560 context lenght where 1000-1500 tokens where used as input text.
|
|
|
36 |
## How to use it:
|
37 |
### Template:
|
38 |
Use the **prompt template** provided below to extract relations from text. Replace `<<your_text_for_extraction>>` with your selected text, ideally between 500-1500 tokens,
|
39 |
+
with an **optimal range** of about **800-1000 tokens**. You can adjust the **temperature** between 0.3 and 1.0; a good starting point is **between 0.6 and 0.7**.
|
40 |
Temperatures below 0.5 may lead to never ending `section_description`. The higher the temperature, the more the model will fill in the gaps in the provided text.
|
41 |
It was **fine-tuned on scientific articles**, so it will supplement missing information with general knowledge.
|
42 |
Model was trained on 2560 context lenght where 1000-1500 tokens where used as input text.
|