Commit
·
a65073c
1
Parent(s):
15c39e7
"Update README.md"
Browse files
README.md
CHANGED
|
@@ -62,7 +62,7 @@ The core project making use of the ggml library is the [llama.cpp](https://githu
|
|
| 62 |
|
| 63 |
There is a bunch of quantized files available. How to choose the best for you:
|
| 64 |
|
| 65 |
-
#
|
| 66 |
|
| 67 |
Q4_0, Q4_1, Q5_0, Q5_1 and Q8 are `legacy` quantization types.
|
| 68 |
Nevertheless, they are fully supported, as there are several circumstances that cause certain model not to be compatible with the modern K-quants.
|
|
@@ -76,6 +76,7 @@ With a Q6_K you should find it really hard to find a quality difference to the o
|
|
| 76 |
|
| 77 |
|
| 78 |
|
|
|
|
| 79 |
# Original Model Card:
|
| 80 |
# Open-Assistant Falcon 7B SFT OASST-TOP1 Model
|
| 81 |
|
|
@@ -192,6 +193,7 @@ python export_model.py --dtype bf16 --hf_repo_name OpenAssistant/falcon-7b-sft-t
|
|
| 192 |
```
|
| 193 |
|
| 194 |
***End of original Model File***
|
|
|
|
| 195 |
|
| 196 |
|
| 197 |
## Please consider to support my work
|
|
|
|
| 62 |
|
| 63 |
There is a bunch of quantized files available. How to choose the best for you:
|
| 64 |
|
| 65 |
+
# Legacy quants
|
| 66 |
|
| 67 |
Q4_0, Q4_1, Q5_0, Q5_1 and Q8 are `legacy` quantization types.
|
| 68 |
Nevertheless, they are fully supported, as there are several circumstances that cause certain model not to be compatible with the modern K-quants.
|
|
|
|
| 76 |
|
| 77 |
|
| 78 |
|
| 79 |
+
---
|
| 80 |
# Original Model Card:
|
| 81 |
# Open-Assistant Falcon 7B SFT OASST-TOP1 Model
|
| 82 |
|
|
|
|
| 193 |
```
|
| 194 |
|
| 195 |
***End of original Model File***
|
| 196 |
+
---
|
| 197 |
|
| 198 |
|
| 199 |
## Please consider to support my work
|