jaro/
llama32_datafusion:latest

7 4 months ago

This model is a fine-tuned version of Llama-3.2-8B-Instruct, specialized for the DataFusion /Rust ecosystem.

... /
params
91b2f57e76f6 · 126B
{
"num_predict": 1024,
"repeat_penalty": 1.2,
"stop": [
"### Instruction:",
"### Response:",
"### End"
],
"temperature": 0.7,
"top_p": 0.9
}