github/gpt4all
1
0
Fork 0
mirror of https://github.com/nomic-ai/gpt4all.git synced 2025-09-06 02:50:36 +00:00
Code Issues Packages Projects Releases Wiki Activity

chat: faster KV shift, continue generating, fix stop sequences (#2781)

Browse Source 
* Don't stop generating at end of context
* Use llama_kv_cache ops to shift context
* Fix and improve reverse prompt detection
* Replace prompt recalc callback with a flag to disallow context shift
This commit is contained in:
Jared Van Bortel
2024-08-07 11:25:24 -04:00
committed by GitHub
parent 90de2d32f8
commit be66ec8ab5
16 changed files with 285 additions and 230 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
gpt4all-chat/chat.cpp
View File

@@ -62,7 +62,7 @@ void Chat::connectLLM()
connect(m_llmodel, &ChatLLM::responseStopped, this, &Chat::responseStopped, Qt::QueuedConnection);
connect(m_llmodel, &ChatLLM::modelLoadingError, this, &Chat::handleModelLoadingError, Qt::QueuedConnection);
connect(m_llmodel, &ChatLLM::modelLoadingWarning, this, &Chat::modelLoadingWarning, Qt::QueuedConnection);
connect(m_llmodel, &ChatLLM::recalcChanged, this, &Chat::handleRecalculating, Qt::QueuedConnection);
connect(m_llmodel, &ChatLLM::restoringFromTextChanged, this, &Chat::handleRestoringFromText, Qt::QueuedConnection);
connect(m_llmodel, &ChatLLM::generatedNameChanged, this, &Chat::generatedNameChanged, Qt::QueuedConnection);
connect(m_llmodel, &ChatLLM::generatedQuestionFinished, this, &Chat::generatedQuestionFinished, Qt::QueuedConnection);
connect(m_llmodel, &ChatLLM::reportSpeed, this, &Chat::handleTokenSpeedChanged, Qt::QueuedConnection);
@@ -252,9 +252,9 @@ void Chat::serverNewPromptResponsePair(const QString &prompt)
m_chatModel->appendResponse("Response: ", prompt);
}
bool Chat::isRecalc() const
bool Chat::restoringFromText() const
{
return m_llmodel->isRecalc();
return m_llmodel->restoringFromText();
}
void Chat::unloadAndDeleteLater()
@@ -320,10 +320,10 @@ void Chat::generatedQuestionFinished(const QString &question)
emit generatedQuestionsChanged();
}
void Chat::handleRecalculating()
void Chat::handleRestoringFromText()
{
Network::globalInstance()->trackChatEvent("recalc_context", { {"length", m_chatModel->count()} });
emit recalcChanged();
emit restoringFromTextChanged();
}
void Chat::handleModelLoadingError(const QString &error)