1055673865
Merge pull request #783 from XXpE3/staging
...
add readme-zh_cn.md
2023-07-26 12:40:09 +03:00
7a4b8abde7
add readme-zh_cn.md
2023-07-26 17:34:22 +08:00
1ad361593f
Merge pull request #779 from ouoertheo/ouoertheo/objective4
...
Objective: Custom Prompts, Current Task Highlight
2023-07-25 16:58:40 +03:00
ae850fdde7
protect default objective prompt, cleanup
2023-07-25 08:45:07 -05:00
c8b3bce8b9
current task highlight/edit prompts
2023-07-25 08:36:27 -05:00
c3d43a7d05
Merge pull request #776 from bdashore3/staging
2023-07-25 09:34:13 +03:00
40f7aa47ad
Generate: Ignore bias addition on continue
...
Continue adds to an existing prompt, so the correct parts of a
message are already present.
Signed-off-by: kingbri <bdashore3@proton.me>
2023-07-24 23:45:53 -04:00
0e45aa7e58
Regex: Fix saving logic
...
- Don't save if the name is empty or undefined
- Warn if the find regex isn't found
- Warn if the "Affects" checkbox isn't selected
This allows for edits to be preserved even if a user incorrectly
changes something with the regex script itself.
Signed-off-by: kingbri <bdashore3@proton.me>
2023-07-24 23:21:54 -04:00
ce3d76d662
Coqui TTS Addition
2023-07-25 09:59:08 +09:00
c4209b9448
Fix naming
2023-07-25 01:13:32 +03:00
83cb3456fc
Run showdown output through sanitize
2023-07-25 00:07:52 +03:00
e15c739e64
Merge branch 'staging' of http://github.com/cohee1207/SillyTavern into staging
2023-07-25 00:01:07 +03:00
0df8c4b6a2
Merge pull request #774 from city-unit/feature/delete
...
More general/functionalized char delete methods
2023-07-24 22:45:41 +03:00
8fdea22379
Move delete calls to functions, generalize delete endpoint.
2023-07-24 15:05:27 -04:00
260bb0af52
Fix opening character with id = '0' in /go and /random commands
2023-07-24 17:22:51 +03:00
cf3baccc5e
Fix spelling mistakes
2023-07-24 11:18:03 +03:00
1bc9bdd64f
Merge pull request #769 from city-unit/feature/docker
2023-07-24 09:34:24 +03:00
b2dc4c6c1b
Add git to Dockerfile
2023-07-24 00:46:07 -04:00
130559d499
#746 Add preset manager for ooba/kobold
2023-07-23 23:52:31 +03:00
06899581db
Update gitignore
2023-07-23 22:10:11 +03:00
e9288874b1
Update RR preset
2023-07-23 22:09:59 +03:00
b6243cdbe1
Merge branch 'release' into staging
2023-07-23 21:19:34 +03:00
ed86e8de04
#765 Fix docker startup
2023-07-23 16:52:25 +03:00
1b9ca4c9d9
Unhide advanced parameters for Novel
2023-07-23 02:09:03 +03:00
dd69fd3934
Auto-scrollheight for SD char prefix
2023-07-23 00:28:23 +03:00
3f47386cba
Adapt GH actions for new branches
2023-07-23 00:28:02 +03:00
83252617be
#612 Character-specific prompt prefixes for SD
2023-07-22 23:57:48 +03:00
a9f8506218
Possible chroma auto-adjust fix
2023-07-22 22:21:37 +03:00
ea68e70ac7
Refine message for dupe confirm
2023-07-22 22:14:16 +03:00
e38a9df538
Confirm character duplication
2023-07-22 21:50:21 +03:00
d3fe20c119
Add editable prompt templates
2023-07-22 21:12:23 +03:00
3feffaabe5
SD refine mode for paintbrush button
2023-07-22 20:32:58 +03:00
52b2c89d64
Merge pull request #764 from toby3d/patch-1
2023-07-22 19:27:33 +03:00
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
80fad96b1f
Fix chat width when switching themes
2023-07-22 18:09:50 +03:00
eccfadb14e
#763 Fix "Continue on Send"
2023-07-22 17:46:59 +03:00
01f1f67de3
Clean-up JB prompt of caret returns and dangling spaces
2023-07-22 16:50:34 +03:00
ef3c8a19fe
Bump package version
2023-07-22 14:19:32 +03:00
732761a427
#724 Fix samplers order being too cool for your tiny phones
2023-07-22 14:18:34 +03:00
212e98302a
Fix empty stat and original param
2023-07-22 14:10:21 +03:00
ed366b15ca
Merge branch 'staging' of http://github.com/cohee1207/SillyTavern into staging
2023-07-22 13:50:22 +03:00
d7b895b92d
Merge pull request #758 from ThisIsPIRI/reppen
2023-07-22 13:10:23 +03:00
0d6346180f
Merge pull request #759 from toby3d/patch-2
2023-07-22 13:09:01 +03:00
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
8e73882c9b
Add rep pen range to ooba settings
2023-07-22 07:56:37 +09:00
e2f1f4a57a
Add repetition penalty range for ooba
2023-07-22 07:38:35 +09:00
7205255d18
Increase default context cap for Kobold and ooba to 4k
2023-07-21 23:47:43 +03:00
bc6e46bac8
Code clean-up
2023-07-21 23:40:53 +03:00
374e931b2e
Don't revert OAI model to default
2023-07-21 23:40:12 +03:00
f28e7cf47c
Redirect to OpenAI site to view usage
2023-07-21 23:35:54 +03:00
Cohee
XXpE3
ouoertheo
kingbri
joe
city-unit
Maxim Lebedev
ThisIsPIRI