Update README.md
Browse files
README.md
CHANGED
|
@@ -4,6 +4,10 @@ license: other
|
|
| 4 |
quantized_by: bartowski
|
| 5 |
---
|
| 6 |
|
|
|
|
|
|
|
|
|
|
|
|
|
| 7 |
## Exllama v2 Quantizations of internlm2-chat-20b-llama-exp
|
| 8 |
|
| 9 |
Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.13">turboderp's ExLlamaV2 v0.0.13</a> for quantization.
|
|
|
|
| 4 |
quantized_by: bartowski
|
| 5 |
---
|
| 6 |
|
| 7 |
+
this quant was made by first converting the model to llama format using https://github.com/InternLM/InternLM/blob/main/tools/convert2llama.py
|
| 8 |
+
|
| 9 |
+
if performance is different than the one converted previously, please comment
|
| 10 |
+
|
| 11 |
## Exllama v2 Quantizations of internlm2-chat-20b-llama-exp
|
| 12 |
|
| 13 |
Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.13">turboderp's ExLlamaV2 v0.0.13</a> for quantization.
|