Update README.md
Browse files
README.md
CHANGED
|
@@ -4,13 +4,19 @@ license: other
|
|
| 4 |
|
| 5 |
This is a GPTQ 4 bit quant of ChanSung's Elina 33b.
|
| 6 |
This is a LlaMa based model; LoRA merged with latest transformers conversion.
|
| 7 |
-
Quantized with GPTQ, --wbits 4 --act-order --true-sequential --save_safetensors
|
|
|
|
| 8 |
128 groupsize was not used so those running this on a consumer GPU with 24GB VRAM can run it at full
|
| 9 |
context (2048) without any risk of OOM.
|
| 10 |
|
|
|
|
| 11 |
Original LoRA:
|
| 12 |
https://huggingface.co/LLMs/Alpaca-LoRA-30B-elina
|
|
|
|
|
|
|
| 13 |
Repo:
|
| 14 |
https://huggingface.co/LLMs
|
|
|
|
|
|
|
| 15 |
Likely Author:
|
| 16 |
https://huggingface.co/chansung
|
|
|
|
| 4 |
|
| 5 |
This is a GPTQ 4 bit quant of ChanSung's Elina 33b.
|
| 6 |
This is a LlaMa based model; LoRA merged with latest transformers conversion.
|
| 7 |
+
Quantized with GPTQ, --wbits 4 --act-order --true-sequential --save_safetensors c4.
|
| 8 |
+
|
| 9 |
128 groupsize was not used so those running this on a consumer GPU with 24GB VRAM can run it at full
|
| 10 |
context (2048) without any risk of OOM.
|
| 11 |
|
| 12 |
+
|
| 13 |
Original LoRA:
|
| 14 |
https://huggingface.co/LLMs/Alpaca-LoRA-30B-elina
|
| 15 |
+
|
| 16 |
+
|
| 17 |
Repo:
|
| 18 |
https://huggingface.co/LLMs
|
| 19 |
+
|
| 20 |
+
|
| 21 |
Likely Author:
|
| 22 |
https://huggingface.co/chansung
|