TinyStories-1Layer-21M-Codebook / eval_results.json
taufeeque's picture
upload model
26ccd30
raw
history blame
484 Bytes
{
"epoch": 10.43,
"eval_MSE/layer0": 6073.8636798095695,
"eval_accuracy": 0.5791389432485323,
"eval_dead_code_fraction/layer0": 0.1981725,
"eval_input_norm/layer0": 0.7182212994247673,
"eval_loss": 1.6604058742523193,
"eval_multicode_k": 1,
"eval_output_norm/layer0": 76.78913438796998,
"eval_runtime": 6.7146,
"eval_samples": 100,
"eval_samples_per_second": 14.893,
"eval_steps_per_second": 0.298,
"perplexity": 5.261445896555633
}