Update README.md
Browse files
README.md
CHANGED
|
@@ -3,6 +3,9 @@ OpenAssistant-Llama-30B-4-bit working with GPTQ versions used in Oobabooga's Tex
|
|
| 3 |
<p>There are 3 quantized versions, one is quantized using GPTQ's <i>--true-sequential</i> and <i>--act-order</i> optimizations, the second is quantized using GPTQ's <i>--true-sequential</i> and <i>--groupsize 128</i> optimization, and the third one is quantized for GGML using q4_1</p>
|
| 4 |
This was made using <a href="https://huggingface.co/OpenAssistant/oasst-sft-7-llama-30b-xor">Open Assistant's native fine-tune</a> of Llama 30b on their dataset.</p>
|
| 5 |
|
|
|
|
|
|
|
|
|
|
| 6 |
<p><strong><font size="5">Update 04.29.2023</font></strong></p>
|
| 7 |
<p>Updated to the latest fine-tune by Open Assistant <a href="https://huggingface.co/OpenAssistant/oasst-sft-7-llama-30b-xor">oasst-sft-7-llama-30b-xor</a>.</p>
|
| 8 |
|
|
|
|
| 3 |
<p>There are 3 quantized versions, one is quantized using GPTQ's <i>--true-sequential</i> and <i>--act-order</i> optimizations, the second is quantized using GPTQ's <i>--true-sequential</i> and <i>--groupsize 128</i> optimization, and the third one is quantized for GGML using q4_1</p>
|
| 4 |
This was made using <a href="https://huggingface.co/OpenAssistant/oasst-sft-7-llama-30b-xor">Open Assistant's native fine-tune</a> of Llama 30b on their dataset.</p>
|
| 5 |
|
| 6 |
+
<p><strong><font size="5">Update 05.19.2023</font></strong></p>
|
| 7 |
+
<p>Updated the ggml quantizations to be compatible with the latest version of llamacpp.</p>
|
| 8 |
+
|
| 9 |
<p><strong><font size="5">Update 04.29.2023</font></strong></p>
|
| 10 |
<p>Updated to the latest fine-tune by Open Assistant <a href="https://huggingface.co/OpenAssistant/oasst-sft-7-llama-30b-xor">oasst-sft-7-llama-30b-xor</a>.</p>
|
| 11 |
|