Update README.md
Browse files
README.md
CHANGED
|
@@ -13,7 +13,8 @@ This is a direct extraction of the 8 experts from [Mixtral-8x7b-Instruct-v0.1](h
|
|
| 13 |
- **Expert Configuration:** It is 2 experts per token.
|
| 14 |
- **Performance:** Performance is identical to instruct, if not a little better.
|
| 15 |
- **Evaluations:** Evals will come, it is more malleable to training.
|
| 16 |
-
- **Experimentation:** This is
|
|
|
|
| 17 |
|
| 18 |
## Instruction Format
|
| 19 |
To leverage instruction fine-tuning, your prompts should be enclosed with `[INST]` and `[/INST]` tokens. The very first instruction should begin with a begin-of-sentence id, while subsequent instructions should not. Assistant generation will conclude with an end-of-sentence token id.
|
|
|
|
| 13 |
- **Expert Configuration:** It is 2 experts per token.
|
| 14 |
- **Performance:** Performance is identical to instruct, if not a little better.
|
| 15 |
- **Evaluations:** Evals will come, it is more malleable to training.
|
| 16 |
+
- **Experimentation:** This is the first of a few MoE expert extraction and modification projects we're working on, more to come.
|
| 17 |
+
- Enjoy.
|
| 18 |
|
| 19 |
## Instruction Format
|
| 20 |
To leverage instruction fine-tuning, your prompts should be enclosed with `[INST]` and `[/INST]` tokens. The very first instruction should begin with a begin-of-sentence id, while subsequent instructions should not. Assistant generation will conclude with an end-of-sentence token id.
|