tingyuansen
commited on
Update README.md
Browse files
README.md
CHANGED
@@ -38,7 +38,7 @@ AstroLLaMA-3-8B-Chat_Summary is a specialized chat model for astronomy, develope
|
|
38 |
- Warmup ratio: 0.03
|
39 |
- Cosine decay schedule for learning rate reduction
|
40 |
- **Primary Use**: Instruction-following and chat-based interactions for astronomy-related queries
|
41 |
-
- **Reference**: Pan et al. 2024
|
42 |
|
43 |
## Using the model for chat
|
44 |
|
@@ -85,7 +85,7 @@ print(f"Assistant: {response}")
|
|
85 |
|
86 |
This model used the summarized content for training, which has led to improved performance compared to the AIC (Abstract, Introduction, Conclusion) version. The summarization process allows for the inclusion of more comprehensive information from each paper while maintaining a manageable token count.
|
87 |
|
88 |
-
Here's a performance comparison chart based upon the astronomical benchmarking Q&A as described in [Ting et al. 2024](https://arxiv.org/abs/2407.11194)
|
89 |
|
90 |
| Model | Score (%) |
|
91 |
|-------|-----------|
|
|
|
38 |
- Warmup ratio: 0.03
|
39 |
- Cosine decay schedule for learning rate reduction
|
40 |
- **Primary Use**: Instruction-following and chat-based interactions for astronomy-related queries
|
41 |
+
- **Reference**: [Pan et al. 2024](https://arxiv.org/abs/2409.19750)
|
42 |
|
43 |
## Using the model for chat
|
44 |
|
|
|
85 |
|
86 |
This model used the summarized content for training, which has led to improved performance compared to the AIC (Abstract, Introduction, Conclusion) version. The summarization process allows for the inclusion of more comprehensive information from each paper while maintaining a manageable token count.
|
87 |
|
88 |
+
Here's a performance comparison chart based upon the astronomical benchmarking Q&A as described in [Ting et al. 2024](https://arxiv.org/abs/2407.11194):
|
89 |
|
90 |
| Model | Score (%) |
|
91 |
|-------|-----------|
|