Qwen2-VL-7B-Instruct-LoRA-FT_16lr / train_results.json
saim1212's picture
second model upload
1d7d53f verified
raw
history blame contribute delete
217 Bytes
{
"epoch": 10.0,
"total_flos": 1.2690688742965248e+16,
"train_loss": 0.4011231598258018,
"train_runtime": 16356.6727,
"train_samples_per_second": 0.061,
"train_steps_per_second": 0.015
}