PEFT
Safetensors
grpo
trl
lora
vision-language-model
topological-reasoning
curvebench
curvebench-gemma-3-12b / train_results.json
AmirMohseni's picture
Model save
fa6814a verified
Raw
History Blame Contribute Delete
201 Bytes
{
"total_flos": 0.0,
"train_loss": 1.7647783374741266e-06,
"train_runtime": 1084.2338,
"train_samples": 40,
"train_samples_per_second": 47.222,
"train_steps_per_second": 0.369
}