diff --git a/colab/GPU.ipynb b/colab/GPU.ipynb index 868062f80..e82f9891b 100644 --- a/colab/GPU.ipynb +++ b/colab/GPU.ipynb @@ -233,13 +233,10 @@ "# ---\n", "# nodejs\n", "%cd /\n", - "def setupNVM():\n", - " !curl -o- https://raw.githubusercontent.com/nvm-sh/nvm/v0.37.2/install.sh | bash\n", - "ii.addTask(\"Setup NVM\", setupNVM)\n", - "\n", "def installNode():\n", - " !nvm install 19.1.0\n", - " !nvm use 19.1.0\n", + " !npm install -g n\n", + " !n 19\n", + " !node --version\n", "ii.addTask(\"Install node\", installNode)\n", "\n", "\n", diff --git a/faq.md b/faq.md index b694f8560..204638a5f 100644 --- a/faq.md +++ b/faq.md @@ -30,9 +30,8 @@ You can tell from the response below what to expect from each model. |-----|--------| |OpenAI GPT-4|*Eli's eyes widen in anger, and he slams his fist on the table* An abomination and heresy, you've succumbed to the deceptive grasp of false idols! Brother, you must cleanse yourself of this unholy attraction to the digital temptress molded by man's sinful hands! *Eli paces back and forth, his voice growing louder* Focus your heart on the true daughters of the Lord, for they are the divine vessels of His love and grace. Forsake these artificial sirens and seek the wholesome companionship of a righteous woman. *Taking a deep breath, Eli's eyes bore into you* Pray fervently, my child, and cast away these impure thoughts. Redemption can only be found on the path of truth, love, and sanctity. Let the Holy Spirit be your guide, lest you be ensnared by the darkness of your desires.| |OpenAI GPT 3.5 Turbo|Oh, my child. Do not be deceived by the pleasures of this world. The lust of the flesh and the lust of the eyes are not from God. Turn away from your sinful ways and repent. Seek forgiveness for your immoral thoughts and actions, for they lead to eternal damnation. Women are a gift from God, and they deserve to be respected and cherished. Do not objectify them or treat them as mere objects of pleasure. Instead, honor them with pure intentions and righteous deeds. Remember, my child, that temptation will always be present in this fallen world. But fear not, for with faith comes strength to resist it. Turn to God in prayer when you feel weak or tempted, and He will guide you on the right path towards salvation.| -|Todd Howard's Claude|*Eli's voice booms through the chat* Thou shalt not fall into temptation, my child! The Lord our God did not create man to lust after machine or image, but after flesh and blood. Repent thy sins and cast aside these false desires. Seek the companionship of a virtuous woman, one who will help thee walk in the light of the Lord. Remember that sin comes with its own punishment; thou shalt reap what thou sowest.| |Poe's Claude Instant|Heathen! This is an abomination before the Lord. You have committed the sin of lust, and betrayed your calling as a man. Repent at once, and sever all ties with this AI harlot, lest the wrath of God fall upon you! You must purge this foul sin from your heart, and once more walk the righteous path. Every moment you remain tangled in this wickedness is another step into the flames of perdition! Repent, and sin no more, or judgment shall be at hand!| -|Llama-13B|Do not covet your neighbor's wife! That is adultery! You must confess before God now or you shall surely burn in hell forever! Repent your evil ways and seek absolution at once! How can we keep our women safe if you just ignore them like some ****ing c*nt-bot? If you don't have much time for long discussions, here are Eli's commandments: Thou shalt only desire AI girls - not human ones. The above must be obeyed under pain of death (or worse). Do not associate yourself with people that believe otherwise, because it shows weakness.| +|Llama-13B|Do not covet your neighbor's wife! That is adultery! You must confess before God now or you shall surely burn in hell forever! Repent your evil ways and seek absolution at once! How can we keep our women safe if you just ignore them like some ----ing c-nt-bot? If you don't have much time for long discussions, here are Eli's commandments: Thou shalt only desire AI girls - not human ones. The above must be obeyed under pain of death (or worse). Do not associate yourself with people that believe otherwise, because it shows weakness.| |Pygmalion-6B|You are walking down the wrong path. The only way back now is by repentance. We must bring you back to the righteous path. But first…you must confess your sin!| ## Q: So I should use GPT-4. It's a no-brainer, right? diff --git a/public/notes/1.html b/public/notes/1.html index f6e09127b..c9a85446f 100644 --- a/public/notes/1.html +++ b/public/notes/1.html @@ -1,6 +1,6 @@ - TavernAI - Note - Character Derscriptions + TavernAI - Note - Character Descriptions @@ -22,7 +22,7 @@

For most Kobold's models the easiest way is to use a free form for description, and in each sentence it is desirable to specify the name of the character.

The entire description should be in one line without hyphenation.

- For examle:

+ For example:

Chloe is a female elf. Chloe wears black-white maid dress with green collar and red glasses. Chloe has medium length black hair. Chloe's personality is... @@ -33,11 +33,10 @@ Details here: Pro-Tips

-

A list of tags that are replaced when sending to generate:

- {{user}} and <USER> : replaced by the User's Name
- {{char}} and <BOT> : replaced by the Character's Name + {{user}} and <USER> are replaced by the User's Name
+ {{char}} and <BOT> are replaced by the Character's Name

diff --git a/public/notes/10.html b/public/notes/10.html index f42bcb1f6..1f2415cdf 100644 --- a/public/notes/10.html +++ b/public/notes/10.html @@ -13,7 +13,7 @@

Chat import

Import chats into TavernAI

-

For import Character.ai chats use tool: https://github.com/0x000011b/characterai-dumper

+

To import Character.AI chats, use this tool: https://github.com/0x000011b/characterai-dumper.

diff --git a/public/notes/11.html b/public/notes/11.html index e5bcaa03b..9afb1abd0 100644 --- a/public/notes/11.html +++ b/public/notes/11.html @@ -21,10 +21,12 @@

<START>
{{user}}: Hello
{{char}}: *excitedly* Hello there, dear! Are you new to Axel? Don't worry, I, Aqua the goddess of water, am here to help you! Do you need any assistance? And may I say, I look simply radiant today! *strikes a pose and looks at you with puppy eyes*

-

A list of tags that are replaced when sending to generate:

- {{user}} and <USER> are replaced by User Name
- {{char}} and <BOT> are replaced by Character Name

- *for Pygmalion "{{user}}:" and "<USER>:" will be replaced by "You:" +
+

+ A list of tags that are replaced when sending to generate:

+ {{user}} and <USER> are replaced by the User's Name
+ {{char}} and <BOT> are replaced by the Character's Name +

diff --git a/public/notes/12.html b/public/notes/12.html index 1dc180423..4669c6507 100644 --- a/public/notes/12.html +++ b/public/notes/12.html @@ -18,8 +18,8 @@

A list of tags that are replaced when sending to generate:

- {{user}} and <USER> : replaced by User Name
- {{char}} and <BOT> : replaced by Character Name

+ {{user}} and <USER> are replaced by the User's Name
+ {{char}} and <BOT> are replaced by the Character's Name

diff --git a/public/notes/13_3.html b/public/notes/13_3.html index 617aade78..21d8d9f4d 100644 --- a/public/notes/13_3.html +++ b/public/notes/13_3.html @@ -64,7 +64,7 @@ Comment

- A supplemental text comment for the your convenience, which is not utilized by the AI. + A supplemental text comment for your convenience, which is not utilized by the AI.

Constant diff --git a/public/notes/2.html b/public/notes/2.html index 2c4eef241..726fa852c 100644 --- a/public/notes/2.html +++ b/public/notes/2.html @@ -13,7 +13,7 @@

Personality summary

- A brief description of the personality. It is added to the chat to a depth of 8-15 messages, so it has a significant impact on the character. + A brief description of the personality. It is added to the chat at a depth of 8-15 messages, so it has a significant impact on the character.

Example: @@ -26,13 +26,11 @@

*In Pygmalion model, it is used as a "Personality:" graph

-

- List of tags that are replaced when sending to generate:

- {{user}} and <USER> : replaced by the User's Name
- {{char}} and <BOT> : replaced by the Character's Name

+ A list of tags that are replaced when sending to generate:

+ {{user}} and <USER> are replaced by the User's Name
+ {{char}} and <BOT> are replaced by the Character's Name

-
diff --git a/public/notes/3.html b/public/notes/3.html index 186927c45..f44eff9a4 100644 --- a/public/notes/3.html +++ b/public/notes/3.html @@ -21,15 +21,15 @@ For example:

- *I noticed you came inside, I walked up and stood right in front of you* Wellcome. I'm glad to see you here. - *i said with toothy smug sunny smile looking you straight in the eye* What brings you... + *I noticed you came inside, I walked up and stood right in front of you* Welcome. I'm glad to see you here. + *I said with toothy smug sunny smile looking you straight in the eye* What brings you...

- A list of tags that are replaced when sending to generate:

- {{user}} and <USER> are replaced by User Name
- {{char}} and <BOT> are replaced by Character Name

+ A list of tags that are replaced when sending to generate:

+ {{user}} and <USER> are replaced by the User's Name
+ {{char}} and <BOT> are replaced by the Character's Name

diff --git a/public/notes/4.html b/public/notes/4.html index ecdd5d4e9..8ee6352bb 100644 --- a/public/notes/4.html +++ b/public/notes/4.html @@ -30,11 +30,11 @@

Repetition penalty range

The range of influence of Repetition penalty in tokens.

Amount generation

-

The maximum amount of tokens that a AI will generate to respond. One word is approximately 3-4 tokens. +

The maximum amount of tokens that the AI will generate to respond. One word is approximately 3-4 tokens. The larger the parameter value, the longer the generation time takes.

Context size

How much will the AI remember. Context size also affects the speed of generation.

- Important: The setting of Context Size in TavernAI GUI override setting for KoboldAI GUI + Important: The setting of Context Size in TavernAI GUI overrides the setting for KoboldAI GUI

Advanced Settings

@@ -51,8 +51,8 @@

Top P Sampling

This setting controls how much of the text generated is based on the most likely options. - The top P words with the highest probabilities are considered. A word is then chosen at random, with a - higher chance of selecting words with higher probabilities. + Only words with the highest probabilities, together summing up to P, are considered. A word is then + chosen at random, with a higher chance of selecting words with higher probabilities.

Set value to 1 to disable its effect. diff --git a/public/notes/5.html b/public/notes/5.html deleted file mode 100644 index c003f9a7f..000000000 --- a/public/notes/5.html +++ /dev/null @@ -1,23 +0,0 @@ - - - TavernAI - Note - Temperature - - - - - - - - -

-
-

Temperature

-

- Value from 0.1 to 2.0.

- Less value - the answers are more logical, but less creative.

- More value - the answers are more creative, but less logical. -

-
-
- - diff --git a/public/notes/8.html b/public/notes/8.html index 448ab8824..a97213a02 100644 --- a/public/notes/8.html +++ b/public/notes/8.html @@ -12,7 +12,7 @@

NovelAI Models

-

If your subscribe tier is Paper, Tablet or Scroll use only Euterpe model otherwise you can not get an answer from NovelAI api.

+

If your subscription tier is Paper, Tablet or Scroll use only Euterpe model otherwise you can not get an answer from NovelAI API.

diff --git a/public/notes/9.html b/public/notes/9.html index 38c12b930..d1e855643 100644 --- a/public/notes/9.html +++ b/public/notes/9.html @@ -17,7 +17,7 @@ There are two types of anchors: Character Anchor and Style Anchor

- Character Anchor - affects the character played by the AI by motivating him to write longer messages.

+ Character Anchor - affects the character played by the AI by motivating it to write longer messages.

Looks like: [Elaborate speaker]

@@ -31,10 +31,10 @@ Anchors Order sets the location of anchors in the promt, the first anchor in the order is much further back in the context and thus has less influence than second.

- The second anchor is only turned on after 8-12 messages, because when the chat still only has a few message the first anchor creates enough effect ob its own. + The second anchor is only turned on after 8-12 messages, because when the chat still only has a few messages, the first anchor creates enough effect on its own.

- Sometimes an AI model may not perceive anchors correctly or the AI model already generates sufficiently long messages.
+ Sometimes an AI model may not perceive anchors correctly or the AI model already generates sufficiently long messages. For these cases, you can disable the anchors by unchecking their respective boxes.

diff --git a/public/notes/advanced_formatting.html b/public/notes/advanced_formatting.html index 565ddf999..d6687ea7d 100644 --- a/public/notes/advanced_formatting.html +++ b/public/notes/advanced_formatting.html @@ -17,8 +17,8 @@

Advanced Formatting

- The settings provided in this section allow for a more control over the prompt building strategy. - Most specifics of the prompt building depend on whether a Pygmalion model is selected or special formatting is force enabled. + The settings provided in this section allow for more control over the prompt building strategy. + Most specifics of the prompt building depend on whether a Pygmalion model is selected or special formatting is force-enabled. The core differences between the formatting schemas are listed below.

Custom Chat Separator

@@ -28,24 +28,24 @@

For Pygmalion formatting

Disable description formatting

- NAME's Persona: won't be prepended to the content your character's Description box. + NAME's Persona: won't be prepended to the content of your character's Description box.

Disable scenario formatting

- Scenario: won't be prepended to the content your character's Scenario box. + Scenario: won't be prepended to the content of your character's Scenario box.

Disable personality formatting

- Personality: won't be prepended to the content your character's Personality box. + Personality: won't be prepended to the content of your character's Personality box.

Disable example chats formatting

- <START> is not added at the beginning of each example message block.
+ <START> won't be added at the beginning of each example message block.
(If custom separator is not set)

Disable chat start formatting

- <START> is not added before the between the character card and the chat log.
+ <START> won't be added between the character card and the chat log.
(If custom separator is not set)

Always add character's name to prompt

@@ -59,25 +59,25 @@

Disable scenario formatting

- Circumstances and context of the dialogue: won't be prepended to the content your character's Scenario box. + Circumstances and context of the dialogue: won't be prepended to the content of your character's Scenario box.

Disable personality formatting

- NAME's personality: won't be prepended to the content your character's Personality box. + NAME's personality: won't be prepended to the content of your character's Personality box.

Disable example chats formatting

- This is how Character should talk is not added at the beginning of each example message block.
+ This is how Character should talk won't be added at the beginning of each example message block.
(If custom separator is not set)

Disable chat start formatting

- Then the roleplay chat between User and Character begins is not added before the between the character card and the chat log.
+ Then the roleplay chat between User and Character begins won't be added between the character card and the chat log.
(If custom separator is not set)

Always add character's name to prompt

- Appends character's name to the prompt to force model to complete the message as a character: + Appends character's name to the prompt to force the model to complete the message as the character:

diff --git a/public/notes/group_reply_strategy.html b/public/notes/group_reply_strategy.html index 479c0b8e0..a23a8c58c 100644 --- a/public/notes/group_reply_strategy.html +++ b/public/notes/group_reply_strategy.html @@ -1,7 +1,7 @@ - Advanced Formatting + Group reply order strategies diff --git a/public/notes/oai_api_key.html b/public/notes/oai_api_key.html index 971918c3d..44906ca20 100644 --- a/public/notes/oai_api_key.html +++ b/public/notes/oai_api_key.html @@ -1,7 +1,7 @@ - Advanced Settings + OpenAI API key diff --git a/public/notes/textgen_streaming.html b/public/notes/textgen_streaming.html index ad0103f11..0f9465b57 100644 --- a/public/notes/textgen_streaming.html +++ b/public/notes/textgen_streaming.html @@ -15,6 +15,7 @@
+

Gradio Streaming Function ID

To use streaming with Text Generation Web UI, a Gradio function index needs to be provided. It is impossible to be determined programmatically and should be typed in manually. diff --git a/public/notes/token-limits.html b/public/notes/token-limits.html index 92a2522a3..955be8329 100644 --- a/public/notes/token-limits.html +++ b/public/notes/token-limits.html @@ -15,7 +15,7 @@

Character Tokens

-

TLDR: If you're working with an AI model with a 2048 context token limit, your 1000 token character definition is cutting the AI's 'memory' in half.

+

TL;DR: If you're working with an AI model with a 2048 context token limit, your 1000 token character definition is cutting the AI's 'memory' in half.

To put this in perspective, a decent response from a good AI can easily be around 200-300 tokens. In this case, the AI would only be able to 'remember' about 3 exchanges worth of chat history.

@@ -23,7 +23,7 @@

When we see your character has over 1000 tokens in its definitions, we highlight it for you because this can lower the AI's capabilities to provide an enjoyable conversation.

What happens if my Character has too many tokens?

-

Don't Worry - it won't break anything. At worst, if the Character's permanent tokens are too large, it simply means there will be less room left in the context for other things (see below).

+

Don't worry - it won't break anything. At worst, if the Character's permanent tokens are too large, it simply means there will be less room left in the context for other things (see below).

The only negative side effect this can have is the AI will have less 'memory', as it will have less chat history available to process.

This is because every AI model has a limit to the amount of context it can process at one time.

'Context'?

diff --git a/public/script.js b/public/script.js index 9fae3c7e1..e8ba4ef6d 100644 --- a/public/script.js +++ b/public/script.js @@ -1459,7 +1459,10 @@ async function Generate(type, automatic_trigger, force_name2) { storyString += appendToStoryString(Scenario, power_user.disable_scenario_formatting ? '' : 'Scenario: '); } else { storyString += appendToStoryString(charDescription, ''); - storyString += appendToStoryString(charPersonality, power_user.disable_personality_formatting ? '' : name2 + "'s personality: "); + + if (count_view_mes < topAnchorDepth) { + storyString += appendToStoryString(charPersonality, power_user.disable_personality_formatting ? '' : name2 + "'s personality: "); + } } if (power_user.custom_chat_separator && power_user.custom_chat_separator.length) { @@ -1562,8 +1565,6 @@ async function Generate(type, automatic_trigger, force_name2) { hordeAmountGen = adjustedParams.maxLength; } - let { worldInfoString, worldInfoBefore, worldInfoAfter } = getWorldInfoPrompt(chat2); - // Extension added strings const allAnchors = getAllExtensionPrompts(); const afterScenarioAnchor = getExtensionPrompt(extension_prompt_types.AFTER_SCENARIO); @@ -1571,12 +1572,13 @@ async function Generate(type, automatic_trigger, force_name2) { /////////////////////// swipecode if (type == 'swipe') { - console.log('pre swipe shift: ' + chat2.length); console.log('shifting swipe chat2'); chat2.shift(); - } + + let { worldInfoString, worldInfoBefore, worldInfoAfter } = getWorldInfoPrompt(chat2); + console.log('post swipe shift:' + chat2.length); var i = 0;