Update README.md
Browse files
README.md
CHANGED
@@ -10,6 +10,77 @@ tags:
|
|
10 |
This model was converted to GGUF format from [`PygmalionAI/Eleusis-12B`](https://huggingface.co/PygmalionAI/Eleusis-12B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
|
11 |
Refer to the [original model card](https://huggingface.co/PygmalionAI/Eleusis-12B) for more details on the model.
|
12 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
13 |
## Use with llama.cpp
|
14 |
Install llama.cpp through brew (works on Mac and Linux)
|
15 |
|
|
|
10 |
This model was converted to GGUF format from [`PygmalionAI/Eleusis-12B`](https://huggingface.co/PygmalionAI/Eleusis-12B) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
|
11 |
Refer to the [original model card](https://huggingface.co/PygmalionAI/Eleusis-12B) for more details on the model.
|
12 |
|
13 |
+
---
|
14 |
+
Alongside the release of Pygmalion-3, we present an additional roleplay model based on Mistral's Nemo Base named Eleusis,
|
15 |
+
a unique model that has a distinct voice among its peers. Though it was
|
16 |
+
meant to be a test run for further experiments, this model was received
|
17 |
+
warmly to the point where we felt it was right to release it publicly.
|
18 |
+
|
19 |
+
|
20 |
+
We release the weights of Eleusis under the Apache 2.0 license, ensuring a free and open ecosystem for it to flourish under.
|
21 |
+
|
22 |
+
|
23 |
+
|
24 |
+
|
25 |
+
|
26 |
+
|
27 |
+
|
28 |
+
Prompting
|
29 |
+
|
30 |
+
|
31 |
+
|
32 |
+
|
33 |
+
Like its component models, Eleusis utilizes the standard ChatML format.
|
34 |
+
|
35 |
+
|
36 |
+
<|im_start|>system
|
37 |
+
Your responses must be detailed, creative, immersive, and drive the scenario forward.<|im_end|>
|
38 |
+
<|im_start|>user
|
39 |
+
{{user}}: Good evening!<|im_end|>
|
40 |
+
<|im_start|>assistant
|
41 |
+
{{char}}:
|
42 |
+
|
43 |
+
|
44 |
+
|
45 |
+
Note that this system prompt is an example and experimentation is encouraged for your use-case purposes. {{user}} and {{char}} are placeholder names and should be replaced with the user's name and the character to be roleplayed as by the model.
|
46 |
+
|
47 |
+
|
48 |
+
|
49 |
+
|
50 |
+
|
51 |
+
|
52 |
+
|
53 |
+
Limitations and biases
|
54 |
+
|
55 |
+
|
56 |
+
|
57 |
+
|
58 |
+
The intended use-case for this model is fictional writing for entertainment purposes. Any other sort of usage is out of scope.
|
59 |
+
|
60 |
+
|
61 |
+
As such, it was not fine-tuned to be safe and
|
62 |
+
harmless: the base model and this fine-tune have been trained on data
|
63 |
+
known to contain profanity and texts that are lewd or otherwise
|
64 |
+
offensive. It may produce socially unacceptable or undesirable text,
|
65 |
+
even if the prompt itself does not include anything explicitly
|
66 |
+
offensive. Outputs might often be factually wrong or misleading.
|
67 |
+
|
68 |
+
|
69 |
+
|
70 |
+
|
71 |
+
|
72 |
+
|
73 |
+
|
74 |
+
Acknowledgements
|
75 |
+
|
76 |
+
|
77 |
+
|
78 |
+
|
79 |
+
A warm thank you is required for the creators of the models we used
|
80 |
+
to construct Eleusis, and a huge shout out once more to Pyg's wonderful
|
81 |
+
community, who's with us every step of the way.
|
82 |
+
|
83 |
+
---
|
84 |
## Use with llama.cpp
|
85 |
Install llama.cpp through brew (works on Mac and Linux)
|
86 |
|