Initial GGML model commit
Browse files
README.md
CHANGED
@@ -19,7 +19,7 @@ license: other
|
|
19 |
|
20 |
# Kaist AI's Selfee 13B GGML
|
21 |
|
22 |
-
These files are GGML format model files for [Kaist AI's Selfee 13B](https://huggingface.co/
|
23 |
|
24 |
GGML files are for CPU + GPU inference using [llama.cpp](https://github.com/ggerganov/llama.cpp) and libraries and UIs which support this format, such as:
|
25 |
* [text-generation-webui](https://github.com/oobabooga/text-generation-webui)
|
@@ -69,16 +69,16 @@ Refer to the Provided Files table below to see what files use which methods, and
|
|
69 |
| selfee-13b.ggmlv3.q3_K_L.bin | q3_K_L | 3 | 6.87 GB | 9.37 GB | New k-quant method. Uses GGML_TYPE_Q5_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K |
|
70 |
| selfee-13b.ggmlv3.q3_K_M.bin | q3_K_M | 3 | 6.25 GB | 8.75 GB | New k-quant method. Uses GGML_TYPE_Q4_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K |
|
71 |
| selfee-13b.ggmlv3.q3_K_S.bin | q3_K_S | 3 | 5.59 GB | 8.09 GB | New k-quant method. Uses GGML_TYPE_Q3_K for all tensors |
|
72 |
-
| selfee-13b.ggmlv3.q4_0.bin | q4_0 | 4 | 7.
|
73 |
-
| selfee-13b.ggmlv3.q4_1.bin | q4_1 | 4 | 8.
|
74 |
| selfee-13b.ggmlv3.q4_K_M.bin | q4_K_M | 4 | 7.82 GB | 10.32 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q4_K |
|
75 |
| selfee-13b.ggmlv3.q4_K_S.bin | q4_K_S | 4 | 7.32 GB | 9.82 GB | New k-quant method. Uses GGML_TYPE_Q4_K for all tensors |
|
76 |
-
| selfee-13b.ggmlv3.q5_0.bin | q5_0 | 5 | 8.
|
77 |
-
| selfee-13b.ggmlv3.q5_1.bin | q5_1 | 5 | 9.
|
78 |
| selfee-13b.ggmlv3.q5_K_M.bin | q5_K_M | 5 | 9.21 GB | 11.71 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q5_K |
|
79 |
| selfee-13b.ggmlv3.q5_K_S.bin | q5_K_S | 5 | 8.95 GB | 11.45 GB | New k-quant method. Uses GGML_TYPE_Q5_K for all tensors |
|
80 |
| selfee-13b.ggmlv3.q6_K.bin | q6_K | 6 | 10.68 GB | 13.18 GB | New k-quant method. Uses GGML_TYPE_Q8_K - 6-bit quantization - for all tensors |
|
81 |
-
| selfee-13b.ggmlv3.q8_0.bin | q8_0 | 8 | 13.
|
82 |
|
83 |
|
84 |
**Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
|
@@ -130,6 +130,63 @@ Thank you to all my generous patrons and donaters!
|
|
130 |
|
131 |
# Original model card: Kaist AI's Selfee 13B
|
132 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
133 |
<p align="center" width="100%">
|
134 |
<a href="https://kaistai.github.io/SelFee/demo" target="_blank"><img src="https://raw.githubusercontent.com/kaistAI/SelFee/main/assets/llama_selfie.png" alt="KAIST-Selfee" style="width: 30%; min-width: 200px; display: block; margin: auto;"></a>
|
135 |
</p>
|
|
|
19 |
|
20 |
# Kaist AI's Selfee 13B GGML
|
21 |
|
22 |
+
These files are GGML format model files for [Kaist AI's Selfee 13B](https://huggingface.co/TheBloke/Selfee-13B-fp16).
|
23 |
|
24 |
GGML files are for CPU + GPU inference using [llama.cpp](https://github.com/ggerganov/llama.cpp) and libraries and UIs which support this format, such as:
|
25 |
* [text-generation-webui](https://github.com/oobabooga/text-generation-webui)
|
|
|
69 |
| selfee-13b.ggmlv3.q3_K_L.bin | q3_K_L | 3 | 6.87 GB | 9.37 GB | New k-quant method. Uses GGML_TYPE_Q5_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K |
|
70 |
| selfee-13b.ggmlv3.q3_K_M.bin | q3_K_M | 3 | 6.25 GB | 8.75 GB | New k-quant method. Uses GGML_TYPE_Q4_K for the attention.wv, attention.wo, and feed_forward.w2 tensors, else GGML_TYPE_Q3_K |
|
71 |
| selfee-13b.ggmlv3.q3_K_S.bin | q3_K_S | 3 | 5.59 GB | 8.09 GB | New k-quant method. Uses GGML_TYPE_Q3_K for all tensors |
|
72 |
+
| selfee-13b.ggmlv3.q4_0.bin | q4_0 | 4 | 7.32 GB | 9.82 GB | Original llama.cpp quant method, 4-bit. |
|
73 |
+
| selfee-13b.ggmlv3.q4_1.bin | q4_1 | 4 | 8.14 GB | 10.64 GB | Original llama.cpp quant method, 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models. |
|
74 |
| selfee-13b.ggmlv3.q4_K_M.bin | q4_K_M | 4 | 7.82 GB | 10.32 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q4_K |
|
75 |
| selfee-13b.ggmlv3.q4_K_S.bin | q4_K_S | 4 | 7.32 GB | 9.82 GB | New k-quant method. Uses GGML_TYPE_Q4_K for all tensors |
|
76 |
+
| selfee-13b.ggmlv3.q5_0.bin | q5_0 | 5 | 8.95 GB | 11.45 GB | Original llama.cpp quant method, 5-bit. Higher accuracy, higher resource usage and slower inference. |
|
77 |
+
| selfee-13b.ggmlv3.q5_1.bin | q5_1 | 5 | 9.76 GB | 12.26 GB | Original llama.cpp quant method, 5-bit. Even higher accuracy, resource usage and slower inference. |
|
78 |
| selfee-13b.ggmlv3.q5_K_M.bin | q5_K_M | 5 | 9.21 GB | 11.71 GB | New k-quant method. Uses GGML_TYPE_Q6_K for half of the attention.wv and feed_forward.w2 tensors, else GGML_TYPE_Q5_K |
|
79 |
| selfee-13b.ggmlv3.q5_K_S.bin | q5_K_S | 5 | 8.95 GB | 11.45 GB | New k-quant method. Uses GGML_TYPE_Q5_K for all tensors |
|
80 |
| selfee-13b.ggmlv3.q6_K.bin | q6_K | 6 | 10.68 GB | 13.18 GB | New k-quant method. Uses GGML_TYPE_Q8_K - 6-bit quantization - for all tensors |
|
81 |
+
| selfee-13b.ggmlv3.q8_0.bin | q8_0 | 8 | 13.83 GB | 16.33 GB | Original llama.cpp quant method, 8-bit. Almost indistinguishable from float16. High resource use and slow. Not recommended for most users. |
|
82 |
|
83 |
|
84 |
**Note**: the above RAM figures assume no GPU offloading. If layers are offloaded to the GPU, this will reduce RAM usage and use VRAM instead.
|
|
|
130 |
|
131 |
# Original model card: Kaist AI's Selfee 13B
|
132 |
|
133 |
+
|
134 |
+
<!-- header start -->
|
135 |
+
<div style="width: 100%;">
|
136 |
+
<img src="https://i.imgur.com/EBdldam.jpg" alt="TheBlokeAI" style="width: 100%; min-width: 400px; display: block; margin: auto;">
|
137 |
+
</div>
|
138 |
+
<div style="display: flex; justify-content: space-between; width: 100%;">
|
139 |
+
<div style="display: flex; flex-direction: column; align-items: flex-start;">
|
140 |
+
<p><a href="https://discord.gg/Jq4vkcDakD">Chat & support: my new Discord server</a></p>
|
141 |
+
</div>
|
142 |
+
<div style="display: flex; flex-direction: column; align-items: flex-end;">
|
143 |
+
<p><a href="https://www.patreon.com/TheBlokeAI">Want to contribute? TheBloke's Patreon page</a></p>
|
144 |
+
</div>
|
145 |
+
</div>
|
146 |
+
<!-- header end -->
|
147 |
+
|
148 |
+
# Kaist AI's Selfee 13B GGML
|
149 |
+
|
150 |
+
This repo contains fp16 pytorch format model files for [Kaist AI's Selfee 13B](https://huggingface.co/kaist-ai/selfee-13b-delta).
|
151 |
+
|
152 |
+
It is the result of merging the diff at the above repo with base Llama 13B, then converting fp32 to fp16.
|
153 |
+
|
154 |
+
## Repositories available
|
155 |
+
|
156 |
+
* [4-bit GPTQ models for GPU inference](https://huggingface.co/TheBloke/Selfee-13B-GPTQ)
|
157 |
+
* [4-bit, 5-bit, and 8-bit GGML models for CPU+GPU inference](https://huggingface.co/TheBloke/Selfee-13B-GGML)
|
158 |
+
* [Unquantised fp16 model in pytorch format, for GPU inference and for further conversions](https://huggingface.co/TheBloke/Selfee-13B-fp16)
|
159 |
+
|
160 |
+
<!-- footer start -->
|
161 |
+
## Discord
|
162 |
+
|
163 |
+
For further support, and discussions on these models and AI in general, join us at:
|
164 |
+
|
165 |
+
[TheBloke AI's Discord server](https://discord.gg/Jq4vkcDakD)
|
166 |
+
|
167 |
+
## Thanks, and how to contribute.
|
168 |
+
|
169 |
+
Thanks to the [chirper.ai](https://chirper.ai) team!
|
170 |
+
|
171 |
+
I've had a lot of people ask if they can contribute. I enjoy providing models and helping people, and would love to be able to spend even more time doing it, as well as expanding into new projects like fine tuning/training.
|
172 |
+
|
173 |
+
If you're able and willing to contribute it will be most gratefully received and will help me to keep providing more models, and to start work on new AI projects.
|
174 |
+
|
175 |
+
Donaters will get priority support on any and all AI/LLM/model questions and requests, access to a private Discord room, plus other benefits.
|
176 |
+
|
177 |
+
* Patreon: https://patreon.com/TheBlokeAI
|
178 |
+
* Ko-Fi: https://ko-fi.com/TheBlokeAI
|
179 |
+
|
180 |
+
**Special thanks to**: Luke from CarbonQuill, Aemon Algiz, Dmitriy Samsonov.
|
181 |
+
|
182 |
+
**Patreon special mentions**: Derek Yates, Sean Connelly, Luke, Nathan LeClaire, Trenton Dambrowitz, Mano Prime, David Flickinger, vamX, Nikolai Manek, senxiiz, Khalefa Al-Ahmad, Illia Dulskyi, trip7s trip, Jonathan Leane, Talal Aujan, Artur Olbinski, Cory Kujawski, Joseph William Delisle, Pyrater, Oscar Rangel, Lone Striker, Luke Pendergrass, Eugene Pentland, Johann-Peter Hartmann.
|
183 |
+
|
184 |
+
Thank you to all my generous patrons and donaters!
|
185 |
+
|
186 |
+
<!-- footer end -->
|
187 |
+
|
188 |
+
# Original model card: Kaist AI's Selfee 13B
|
189 |
+
|
190 |
<p align="center" width="100%">
|
191 |
<a href="https://kaistai.github.io/SelFee/demo" target="_blank"><img src="https://raw.githubusercontent.com/kaistAI/SelFee/main/assets/llama_selfie.png" alt="KAIST-Selfee" style="width: 30%; min-width: 200px; display: block; margin: auto;"></a>
|
192 |
</p>
|