From 6e5ff5f9d7977e40b051c6ab7c95af867136d1c3 Mon Sep 17 00:00:00 2001 From: Cohee <18619528+Cohee1207@users.noreply.github.com> Date: Fri, 22 Sep 2023 23:36:05 +0300 Subject: [PATCH] Update advancedformatting.md --- Usage/Core Concepts/advancedformatting.md | 26 ----------------------- 1 file changed, 26 deletions(-) diff --git a/Usage/Core Concepts/advancedformatting.md b/Usage/Core Concepts/advancedformatting.md index 1ebfb90e8..9f1e38efd 100644 --- a/Usage/Core Concepts/advancedformatting.md +++ b/Usage/Core Concepts/advancedformatting.md @@ -94,29 +94,3 @@ Supported APIs: 2. oobabooga's Text Generation WebUI 3. NovelAI 4. OpenAI, including via OpenRouter (max 4 strings) - -### Multigen - -*This feature provides a pseudo-streaming functionality that conflicts with token streaming. When Multigen is enabled and generation API supports streaming, only Multigen streaming will be used.* - -SillyTavern tries to create faster and longer responses by chaining the generation using smaller batches. - -#### Default settings - -First batch = 50 tokens - -Next batches = 30 tokens - -#### Algorithm - -1. Generate the first batch (if the amount of generation setting is more than batch length). -2. Generate the next batch of tokens until one of the stopping conditions is reached. -3. Append the generated text to the next cycle's prompt. - -#### Stopping conditions - -1. Generated enough text. -2. The character starts speaking for You. -3. <|endoftext|> token reached. -4. No text generated. -5. Stop sequence generated. (Instruct mode only)