llama3-1b-coding-gpt4o-100k2 / train_results.json
chansung's picture
Model save
3888f2b verified
raw
history blame contribute delete
240 Bytes
{
"epoch": 10.0,
"total_flos": 1.0444708917333197e+18,
"train_loss": 1.0473914388348073,
"train_runtime": 1591.5084,
"train_samples": 116368,
"train_samples_per_second": 108.407,
"train_steps_per_second": 0.214
}