aws-kh commited on
Commit ·
a608366
1
Parent(s): 26912c3
corrected table formatting error for topic retrieval results
Browse files
README.md
CHANGED
|
@@ -26,9 +26,10 @@ Although the performance of the models on long context was fairly competitive on
|
|
| 26 |
there were some limitations on its performance on longer context. Motivated by improving its performance on longer context, we finetuned the Mistral 7B model, and produced `Mistrallite`. The model managed to `significantly boost the performance of long context handling` over Mistral-7B-Instruct-v0.1. The detailed `long context evalutaion results` are as below:
|
| 27 |
|
| 28 |
1. [Topic Retrieval](https://lmsys.org/blog/2023-06-29-longchat/)
|
|
|
|
| 29 |
|Model Name|Input length| Input length | Input length| Input length| Input length|
|
| 30 |
|----------|-------------:|-------------:|------------:|-----------:|-----------:|
|
| 31 |
-
| | 2851| 5568 |8313 | 11044 | 13780
|
| 32 |
| Mistral-7B-Instruct-v0.1 | 100% | 50% | 2% | 0% | 0% |
|
| 33 |
| MistralLite | **100%** | **100%** | **100%** | **100%** | **98%** |
|
| 34 |
|
|
|
|
| 26 |
there were some limitations on its performance on longer context. Motivated by improving its performance on longer context, we finetuned the Mistral 7B model, and produced `Mistrallite`. The model managed to `significantly boost the performance of long context handling` over Mistral-7B-Instruct-v0.1. The detailed `long context evalutaion results` are as below:
|
| 27 |
|
| 28 |
1. [Topic Retrieval](https://lmsys.org/blog/2023-06-29-longchat/)
|
| 29 |
+
|
| 30 |
|Model Name|Input length| Input length | Input length| Input length| Input length|
|
| 31 |
|----------|-------------:|-------------:|------------:|-----------:|-----------:|
|
| 32 |
+
| | 2851| 5568 |8313 | 11044 | 13780 |
|
| 33 |
| Mistral-7B-Instruct-v0.1 | 100% | 50% | 2% | 0% | 0% |
|
| 34 |
| MistralLite | **100%** | **100%** | **100%** | **100%** | **98%** |
|
| 35 |
|