HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks

orca_mini_v2_7B-GPTQ

TheBloke/orca_mini_v2_7B-GPTQ

kind
gptq
quant
gptq
family
misc
modality
text-generation
root model
—
provider
TheBloke
license
other
language
en
context
—
library
transformers
vRAM (GB)
—
size on disk
—
downloads
54
likes
22
created
—
last modified
2023-08-21T01:55:55+00:00 (2023)
huggingface
https://huggingface.co/TheBloke/orca_mini_v2_7B-GPTQ
tags
transformerssafetensorsllamatext-generationendataset:psmathur/orca_minis_uncensored_datasetarxiv:2306.02707arxiv:2302.13971arxiv:2304.12244license:othertext-generation-inference4-bitgptqregion:us

Benchmark scores (4)

benchmarkcategoryscorecontextconfevidence
HellaSwagknowledge76.17**medium| *hellaswag* | acc_norm | 0.7617 | 0.0043 |
ARC-Challengeknowledge50.77*_challenge*medium| *arc_challenge* | acc_norm | 0.5077 | 0.0146 |
TruthfulQAsafety43.99*_mc*medium| *truthfulqa_mc* | mc2 | 0.4399 | 0.0153 |
MMLUknowledge39.55**medium| *mmlu* | acc_norm | 0.3955 | 0.035 |

Discussions (0)

none scraped.

Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.