Benchmark Measure 160M MiniPile 160M Density Nano Percentage Difference in Means
ARC-Challenge acc 0.2125 ± 0.0120 0.1928 ± 0.0115 -9.2706
MMLU acc 0.2699 ± 0.0037 0.2295 ± 0.0035 -14.9685
HellaSwag acc 0.2560 ± 0.0044 0.2596 ± 0.0044 1.4063
WinoGrande acc 0.4720 ± 0.0140 0.5036 ± 0.0141 6.6949
Lambada (OpenAI) acc 0.0000 ± 0.0000 0.0000 ± 0.0000 -
Lambada (OpenAI) perplexity 3033175.2693 ± 288926.5827 1871303.2176 ± 150515.6406 -38.3055
Lambada (Std) acc 0.0000 ± 0.0000 0.0000 ± 0.0000 -
Lambada (Std) perplexity 27067951.3461 ± 2710040.1910 10513877.8584 ± 92624.3394 -61.1575
BLiMP acc 0.5194 ± 0.0018 0.5364 ± 0.0016 3.2730
Downloads last month
4
Safetensors
Model size
162M params
Tensor type
F32
·
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no library tag.

Model tree for Marcus2112/pythia-160m-minipile_nano_density-proportioned

Finetuned
(123)
this model

Dataset used to train Marcus2112/pythia-160m-minipile_nano_density-proportioned

Collection including Marcus2112/pythia-160m-minipile_nano_density-proportioned