CodeStral-RAG-Lora / Codestral-22B_Lora /training_prompt.json
TroyDoesAI's picture
An early checkpoint ~ Contextually obedient in answering, training curriculum is currently at training for contextual-refusals when the information provided is not sufficent to answer
a34c2f6 verified
raw
history blame contribute delete
312 Bytes
{
"template_type": "dataset",
"template_1": "Code-Assistant-Request: %instruction%\n\n### Code:\n%output%",
"template_2": "Mermaid-Generator-Request: %instruction%\n\n### Input:\n%input%\n\n### Mermaid:\n%output%",
"template_3": "Contextual-Request: %prompt%\n\n### Contextual Response:\n%chosen%"
}