Update README.md
Browse files
README.md
CHANGED
|
@@ -14,6 +14,8 @@ quantized_by: bartowski
|
|
| 14 |
pipeline_tag: text-generation
|
| 15 |
---
|
| 16 |
|
|
|
|
|
|
|
| 17 |
## Exllama v2 Quantizations of dolphin-2.6.1-mixtral-8x7b
|
| 18 |
|
| 19 |
Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.11">turboderp's ExLlamaV2 v0.0.11</a> for quantization.
|
|
@@ -24,7 +26,7 @@ Conversion was done using the default calibration dataset.
|
|
| 24 |
|
| 25 |
Default arguments used except when the bits per weight is above 6.0, at that point the lm_head layer is quantized at 8 bits per weight instead of the default 6.
|
| 26 |
|
| 27 |
-
Original model: https://huggingface.co/cognitivecomputations/dolphin-2.6.1-mixtral-8x7b
|
| 28 |
|
| 29 |
<a href="https://huggingface.co/bartowski/dolphin-2.6.1-mixtral-8x7b-exl2/tree/3_0">3.0 bits per weight</a>
|
| 30 |
|
|
|
|
| 14 |
pipeline_tag: text-generation
|
| 15 |
---
|
| 16 |
|
| 17 |
+
# Eric has pulled this model due to decreased performance, will leave the quants up but downloader beware, performance isn't what was expected
|
| 18 |
+
|
| 19 |
## Exllama v2 Quantizations of dolphin-2.6.1-mixtral-8x7b
|
| 20 |
|
| 21 |
Using <a href="https://github.com/turboderp/exllamav2/releases/tag/v0.0.11">turboderp's ExLlamaV2 v0.0.11</a> for quantization.
|
|
|
|
| 26 |
|
| 27 |
Default arguments used except when the bits per weight is above 6.0, at that point the lm_head layer is quantized at 8 bits per weight instead of the default 6.
|
| 28 |
|
| 29 |
+
Original model: ~https://huggingface.co/cognitivecomputations/dolphin-2.6.1-mixtral-8x7b~
|
| 30 |
|
| 31 |
<a href="https://huggingface.co/bartowski/dolphin-2.6.1-mixtral-8x7b-exl2/tree/3_0">3.0 bits per weight</a>
|
| 32 |
|