qwen2-0.5b-sft / train_results.json
hZzy's picture
Model save
051a63e verified
raw
history blame contribute delete
248 Bytes
{
"epoch": 0.9993049349617714,
"total_flos": 106140763422720.0,
"train_loss": 1.5477893754295022,
"train_runtime": 7899.7649,
"train_samples": 207864,
"train_samples_per_second": 30.594,
"train_steps_per_second": 0.159
}