Fixed a mistake in quantization table (#3)

- Fixed a mistake in quantization table (8d257074f9699016d713d26dfa15084ce98667c7)


Co-authored-by: Nils <[email protected]>

README.md

@@ -46,7 +46,7 @@ The files in the `previous_llama_ggmlv2` branch are from Epoch 1.
 | Name | Quant method | Bits | Size | RAM required | Use case |
 | ---- | ---- | ---- | ---- | ---- | ----- |
 `manticore-13B.ggmlv3.q4_0.bin` | q4_0 | 4bit | 8.14GB | 10.5GB | 4-bit. |
-`manticore-13B.ggmlv3.q4_1.bin` | q4_0 | 4bit | 8.14GB | 10.5GB | 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
+`manticore-13B.ggmlv3.q4_1.bin` | q4_1 | 4bit | 8.14GB | 10.5GB | 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
 `manticore-13B.ggmlv3.q5_0.bin` | q5_0 | 5bit | 8.95GB | 11.0GB | 5-bit. Higher accuracy, higher resource usage and slower inference. |
 `manticore-13B.ggmlv3.q5_1.bin` | q5_1 | 5bit | 9.76GB | 12.25GB | 5-bit. Even higher accuracy, and higher resource usage and slower inference. |
 `manticore-13B.ggmlv3.q8_0.bin` | q8_0 | 8bit | 14.6GB | 17GB | 8-bit. Almost indistinguishable from float16. Huge resource use and slow. Not recommended for normal use. |