PyTorch
llama
alignment-handbook
Generated from Trainer
Mamba2InLlama_0_75 / train_results.json
Junxiong Wang
add models
2804acb
{
"epoch": 1.0,
"total_flos": 0.0,
"train_loss": 0.5209795107882678,
"train_runtime": 14376.1789,
"train_samples": 133368,
"train_samples_per_second": 9.277,
"train_steps_per_second": 0.29
}