Update README.md
Browse files
README.md
CHANGED
|
@@ -10,6 +10,10 @@ license: llama3
|
|
| 10 |
|
| 11 |
License: https://llama.meta.com/llama3/license
|
| 12 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
240421: Bellman's back!
|
| 14 |
This is a preliminary test run, run on mostly the same settings and dataset as bellman-mistral-instruct (But not the dpo pass). Context length is 3072.
|
| 15 |
I've done some basic testing, and it's not a total mess. Whether it's an improvement to llama-3-instruct, I'm not sure, because that's REALLY good.
|
|
|
|
| 10 |
|
| 11 |
License: https://llama.meta.com/llama3/license
|
| 12 |
|
| 13 |
+
240508: Where is the real version?
|
| 14 |
+
|
| 15 |
+
There's been a number of issues with llama 3 finetuning, and most recently an issue with the tokenizer for ggufs. I'm holding back any release until the situation has stabilizied.
|
| 16 |
+
|
| 17 |
240421: Bellman's back!
|
| 18 |
This is a preliminary test run, run on mostly the same settings and dataset as bellman-mistral-instruct (But not the dpo pass). Context length is 3072.
|
| 19 |
I've done some basic testing, and it's not a total mess. Whether it's an improvement to llama-3-instruct, I'm not sure, because that's REALLY good.
|