Commit
•
ebf85fc
1
Parent(s):
26a8b94
Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,11 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
datasets:
|
| 3 |
+
- EleutherAI/wikitext_document_level
|
| 4 |
+
tags:
|
| 5 |
+
- llama
|
| 6 |
+
---
|
| 7 |
+
LLaMA 33b finetuned on `wikitext_document_level` with combined linear and NTK-aware ROPE scaling (alpha=4, scale=2.)
|
| 8 |
+
This model will be coherent up to at least 8k context length, but might work beyond that.
|
| 9 |
+
This is a merged version of [llama33b-s2a4-qlora](https://huggingface.co/chargoddard/llama33b-s2a4-qlora).
|
| 10 |
+
|
| 11 |
+
Note that this is *not* an instruct model - this is base LLaMA with an extended sequence length.
|