argumentation-prediction / post-training eval_results.json
Alex Spangher
updated
0219ce0
raw
history blame
202 Bytes
{
"eval_samples": 34,
"test_f1": 0.6595240802735967,
"test_loss": 0.6290570497512817,
"test_runtime": 2.9031,
"test_samples_per_second": 11.712,
"test_steps_per_second": 11.712
}