gemma7b-gpt4o_100k_summarize-k / train_results.json
chansung's picture
Model save
fa9e110 verified
{
"epoch": 0.9996810207336523,
"total_flos": 5.97257971054936e+17,
"train_loss": 2.314191315838997,
"train_runtime": 7526.9596,
"train_samples": 115376,
"train_samples_per_second": 1.666,
"train_steps_per_second": 0.208
}