gemma2b-summarize-gpt4o-64k / train_results.json
chansung's picture
Model save
2d3c737 verified
raw
history blame
237 Bytes
{
"epoch": 15.0,
"total_flos": 1.2863476116823736e+18,
"train_loss": 1.080002195214572,
"train_runtime": 11705.6736,
"train_samples": 64610,
"train_samples_per_second": 8.973,
"train_steps_per_second": 0.187
}