Update README.md
Browse files
README.md
CHANGED
|
@@ -43,8 +43,6 @@ it contains the following quantizations of the original weights from Together's
|
|
| 43 |
* [Q8_0](https://huggingface.co/rozek/LLaMA-2-7B-32K-Instruct_GGUF/blob/main/LLaMA-2-7B-32K-Instruct-Q8_0.gguf) and
|
| 44 |
* [F16](https://huggingface.co/rozek/LLaMA-2-7B-32K-Instruct_GGUF/blob/main/LLaMA-2-7B-32K-Instruct-f16.gguf) (unquantized)
|
| 45 |
|
| 46 |
-
(strikethrough files are currently being uploaded)
|
| 47 |
-
|
| 48 |
> Nota bene: while RoPE makes inferences with large contexts possible, you still need an awful lot of RAM
|
| 49 |
> when doing so. And since "32K" does not mean that you always have to use a context size of 32768 (only that
|
| 50 |
> the model was fine-tuned for that size), it is recommended that you keep your context as small as possible
|
|
|
|
| 43 |
* [Q8_0](https://huggingface.co/rozek/LLaMA-2-7B-32K-Instruct_GGUF/blob/main/LLaMA-2-7B-32K-Instruct-Q8_0.gguf) and
|
| 44 |
* [F16](https://huggingface.co/rozek/LLaMA-2-7B-32K-Instruct_GGUF/blob/main/LLaMA-2-7B-32K-Instruct-f16.gguf) (unquantized)
|
| 45 |
|
|
|
|
|
|
|
| 46 |
> Nota bene: while RoPE makes inferences with large contexts possible, you still need an awful lot of RAM
|
| 47 |
> when doing so. And since "32K" does not mean that you always have to use a context size of 32768 (only that
|
| 48 |
> the model was fine-tuned for that size), it is recommended that you keep your context as small as possible
|