twnvy/ edge-lm:latest

21 5 months ago

A token-efficient thinking model that rivals the CoT of much larger models with only 3 billion parameters.

thinking
... /
params
f43dca5c1f16 · 100B
{
"num_ctx": 4096,
"stop": [
"<|eot_id|>",
"<|start_header_id|>"
],
"temperature": 0.15
}