model_007-70B-GGUF

Home/Models/TheBloke/model_007-70B-GGUF

TheBloke/model_007-70B-GGUF

kind: gguf
quant: gguf (2-bit)
family: misc
modality: —
root model: psmathur/model_007
provider: TheBloke
license: llama2
language: en
context: —
library: transformers
vRAM (GB): —
size on disk: —
downloads: 449
likes: 9
created: —
last modified: 2023-09-27T12:46:39+00:00 (2023)
huggingface: https://huggingface.co/TheBloke/model_007-70B-GGUF
tags: transformersggufllamaenarxiv:2306.02707base_model:pankajmathur/model_007base_model:quantized:pankajmathur/model_007license:llama2region:us

benchmark	category	score	context	conf	evidence
HellaSwag	knowledge	87.65	**	medium	`\| hellaswag \| acc_norm \| 0.8765 \| 0.0038 \|`
ARC-Challenge	knowledge	71.08	_challenge	medium	`\| arc_challenge \| acc_norm \| 0.7108 \| 0.0141 \|`
MMLU	knowledge	69.04	**	medium	`\| mmlu \| acc_norm \| 0.6904 \| 0.0351 \|`
TruthfulQA	safety	63.12	_mc	medium	`\| truthfulqa_mc \| mc2 \| 0.6312 \| 0.0157 \|`

none scraped.