llama3-8b-lora-coding-11-v1 / train_results.json
chansung's picture
Model save
7c9ca48 verified
raw
history blame
236 Bytes
{
"epoch": 1.0,
"total_flos": 8.08957492854784e+17,
"train_loss": 1.2461711733880705,
"train_runtime": 2338.1563,
"train_samples": 116368,
"train_samples_per_second": 14.952,
"train_steps_per_second": 0.059
}