Commit Graph

35 Commits

Author SHA1 Message Date
Cohee
af7c89678c Add .gitkeeps 2024-03-28 22:36:18 +02:00
DreamGenX
5c410986a4 Add support for DreamGen API.
API docs: https://dreamgen.com/docs/models/opus/v1
API keys: https://dreamgen.com/account/api-keys

I decided to base this on the text-completion API since it's more
flexible with SillyTavern's prompt formating capabilities.

This also means that custom context and instruct settings are required.

Will add documentation in a followup PR.
2024-03-07 12:25:48 +01:00
Cohee
f669b959c3 Rename RP instruct 2024-02-02 17:37:38 +02:00
Paniphon
9fb631b2f1
Add Alpaca-Single-Turn instruct template 2023-12-29 01:23:31 +07:00
SDS
d9fc396b39
Update Llama 2 Chat.json
Official separator is a space instead of a newline. That format itself is still FUBAR, but at least we get a little closer to what it was intended to look, I guess.
2023-12-19 00:48:42 +01:00
Cohee
4fbbd34db9 Add presets for Story and Adventure 2023-10-07 23:18:00 +03:00
SDS
5848ec498b
Assorted fixes and improvements (#1208)
* Kobold Presets fixed

* Help texts fixed

* Scale API for connectAPISlash

* Quick Reply checkboxes fixed

* New Instruct Mode Presets

* More date/time macros

* ChatML context template and instruct prompt format

* Mistral context template and instruct prompt format

* Removed use_default_badwordsids from kobold presets

* Renamed ChatML to Mistral-OpenOrca (ChatML)

* Renamed Mistral-OpenOrca (removed ChatML)

* Removed single_line from kobold presets

* Removed obsolete use_stop_sequence setting

* Ban EOS Token off by default

* Split AI Resp. Conf. in global and preset-specific settings

* Resolve conflicts

* Fix title

* Add translations for new help texts

* Fix i18n.json whitespace

* Make Mistral-OpenOrca system prompt more generic

* Renamed "Mistral-OpenOrca" to "ChatML" again

* More (UI) fixes and improvements

* Sendas hint fixed

---------

Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
2023-10-07 19:25:36 +03:00
majick
368cba1149
Add Lightning 1.1 Context and Instruct templates (#1214)
* Add Lightning 1.1 Context and Instruct templates

* Remove double white space

* Fix fancy apostrophe

---------

Co-authored-by: Cohee <18619528+Cohee1207@users.noreply.github.com>
2023-10-03 20:05:15 +03:00
Cohee
af4f60a4af Update Pygmalion context template. Add Pygmalion instruct template 2023-09-06 17:56:49 +03:00
Stefan Daniel Schwarz
3e0ce12b23 first_output_sequence and system_sequence_prefix 2023-08-25 22:34:08 +02:00
Stefan Daniel Schwarz
be3ebcda0d improved instruct mode sequences (WIP) 2023-08-25 14:01:25 +02:00
Stefan Daniel Schwarz
56a6398189 system prompt in system sequence 2023-08-24 19:33:04 +02:00
Stefan Daniel Schwarz
582464a2e7 fix name 2023-08-24 18:47:15 +02:00
Cohee
8fea486e57 #1009 Update Vicuna 1.1 template 2023-08-24 15:50:14 +03:00
Stefan Daniel Schwarz
dd7b21c63d renamed roleplay instruct preset 2023-08-23 22:23:51 +02:00
Stefan Daniel Schwarz
9b932dfa15 byebye wizard 2023-08-23 20:24:45 +02:00
Stefan Daniel Schwarz
252be20c16 Return of the Roleplay Context 2023-08-22 23:40:47 +02:00
Stefan Daniel Schwarz
c6ce06b339 Put "### Input:" back into Roleplay system prompt 2023-08-21 23:24:24 +02:00
Stefan Daniel Schwarz
9df4c51b07 Instruct Mode improvements 2023-08-21 22:32:58 +02:00
Cohee
897ad28477 Apply preset manager for instruct presets 2023-08-21 00:56:53 +03:00
Stefan Daniel Schwarz
d931fc8d37 Llama 2 instruct mode preset fixed 2023-08-20 22:04:11 +02:00
Stefan Daniel Schwarz
7b8d10d25e Added Include Names and Forced Names to all instruct presets 2023-08-20 16:10:22 +02:00
Stefan Daniel Schwarz
2fb86c8ebd Added Persona Description and System Prompt to Story String 2023-08-20 16:08:54 +02:00
Stefan Daniel Schwarz
357b765897 roleplay preset newlines 2023-08-20 01:03:34 +02:00
Stefan Daniel Schwarz
a40df0c117 Added simple-proxy-for-tavern context template and instruct preset 2023-08-18 21:16:04 +02:00
Stefan Daniel Schwarz
6a89967144 Added Roleplay context template and updated instruct preset 2023-08-18 21:15:44 +02:00
Cohee
c24f636a10 Add Last output sequence setting for instruct mode 2023-08-17 23:33:27 +03:00
SDS
ef9d40428e
Update Roleplay instruct mode preset: Include Names
After extensive testing, I've enabled "Include Names" to fix some models talking/acting as User.
2023-08-13 20:18:22 +02:00
Stefan Daniel Schwarz
bfac10028f removed enabled field from presets 2023-08-04 23:52:36 +02:00
SDS
45a66fae9d
Added OpenOrca-OpenChat instruct preset
Here's an Instruct Mode preset for OpenOrca-OpenChat, e. g. for [Open-Orca/OpenOrcaxOpenChat-Preview2-13B](https://huggingface.co/Open-Orca/OpenOrcaxOpenChat-Preview2-13B) and other OpenOrca/OpenChat models
2023-08-04 21:44:47 +02:00
SDS
c57de3d47b
Improved Roleplay Preset
More like simple-proxy-for-tavern's default verbose prompt, and now works with SillyTavern's AutoFormat Overrides on or off, so this new and improved version is better and more compatible.
2023-08-02 17:42:42 +02:00
SDS
0c0e24323c proxy-replacement-preset: Roleplay Instruct Mode Preset
In [[Feature Request] Add Simple Proxy functionality into Silly Tavern directly · Issue #831 · SillyTavern/SillyTavern](https://github.com/SillyTavern/SillyTavern/issues/831), I explained how to replace the [simple-proxy-for-tavern](https://github.com/anon998/simple-proxy-for-tavern) using built-in functionality of SillyTavern. To make this easier, here's an Instruct Mode Preset that helps setting this up.
2023-08-01 23:14:50 +03:00
Maxim Lebedev
59100a0076
Fixed tags order in Llama2 instruct preset
I didn't notice that the prompt wording implies impersonation, which is incorrect for common usage. Changed the order of tags used to generate the character's response to the user, rather than the vice versa.
2023-07-22 21:31:44 +06:00
Maxim Lebedev
433ddf5901
Created Llama2 instruct preset
Tested on Llama 2 13B chat model provided by TheBloke. Based on these prompt example: https://huggingface.co/TheBloke/Llama-2-13B-chat-GPTQ#prompt-template-llama-2-chat
2023-07-22 13:24:17 +06:00
Cohee
edd41989fd Initial commit 2023-07-20 20:32:15 +03:00