mkchaou/ calm_q3_32b_r16_a32_lr2e-5_e15:latest

11 2 weeks ago

This model is Qwen3-32b-Instruct fine-tuned on ~13k Climsight data which are pairs of prompts/completions extracted from climsight pipeline. Below are all the technical specs of the fine-tuning process along with the evaluation used.

tools
... /
params
67ed3379cd1b · 181B
{
"num_ctx": 24576,
"stop": [
"<|im_start|>",
"<|im_end|>",
"<|endoftext|>",
"</think>",
"<tool_response>"
],
"temperature": 0.6,
"top_p": 0.95
}