Qwen2.5-7B-CLIPPER / train_results.json
chtmp223's picture
Uploaded all model files (excluding checkpoints)
d3b5399 verified
raw
history blame
206 Bytes
{
"epoch": 1.0,
"total_flos": 3248542671962112.0,
"train_loss": 0.47701127381671354,
"train_runtime": 533164.508,
"train_samples_per_second": 0.186,
"train_steps_per_second": 0.012
}