Update context length
Browse files
README.md
CHANGED
@@ -5,7 +5,7 @@ license_link: https://huggingface.co/microsoft/Orca-2-13b/blob/main/LICENSE
|
|
5 |
pipeline_tag: text-generation
|
6 |
---
|
7 |
|
8 |
-
# OrcaMaid-13b-v2-FIX-
|
9 |
|
10 |
This is the fixed version of **OrcaMaid-v2-13b**, further extended to 10240 context length via YaRN. The (now-deleted) v2 model had issues with the merged tokenizer that prevented it from stopping when necessary, and caused it to output broken ChatML tokens like `<|im_end`, etc.
|
11 |
|
@@ -31,7 +31,7 @@ Below is an instruction that describes a task. Write a response that appropriate
|
|
31 |
### Misc. information
|
32 |
- BOS token is `<s>`
|
33 |
- EOS token is `</s>`
|
34 |
-
- Native context length is `
|
35 |
- Base model is Llama 2
|
36 |
- Due to the inclusion of Orca-2-13b, the model is subject to the terms of the [Microsoft Research License](https://huggingface.co/microsoft/Orca-2-13b/blob/main/LICENSE)
|
37 |
|
|
|
5 |
pipeline_tag: text-generation
|
6 |
---
|
7 |
|
8 |
+
# OrcaMaid-13b-v2-FIX-32k
|
9 |
|
10 |
This is the fixed version of **OrcaMaid-v2-13b**, further extended to 10240 context length via YaRN. The (now-deleted) v2 model had issues with the merged tokenizer that prevented it from stopping when necessary, and caused it to output broken ChatML tokens like `<|im_end`, etc.
|
11 |
|
|
|
31 |
### Misc. information
|
32 |
- BOS token is `<s>`
|
33 |
- EOS token is `</s>`
|
34 |
+
- Native context length is `32768` via YaRN (original context length was `4096`)
|
35 |
- Base model is Llama 2
|
36 |
- Due to the inclusion of Orca-2-13b, the model is subject to the terms of the [Microsoft Research License](https://huggingface.co/microsoft/Orca-2-13b/blob/main/LICENSE)
|
37 |
|