PyTorch
llama
alignment-handbook
Generated from Trainer
MambaInLlama_0_75 / train_results.json
Junxiong Wang
add models
93df259
raw
history blame
220 Bytes
{
"epoch": 1.0,
"total_flos": 0.0,
"train_loss": 0.49609584714538074,
"train_runtime": 16148.8615,
"train_samples": 133368,
"train_samples_per_second": 8.259,
"train_steps_per_second": 0.258
}