zephyr-7b-Cal-DPO-1-001-False / train_results.json
Teng Xiao
TX
87641fc
raw
history blame
228 Bytes
{
"epoch": 0.998691442030882,
"total_flos": 0.0,
"train_loss": 114854.6911687631,
"train_runtime": 7479.76,
"train_samples": 61135,
"train_samples_per_second": 8.173,
"train_steps_per_second": 0.064
}