HF Model & Benchmark Matrix

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

open-instruct-unnatural-instructions-7b

allenai/open-instruct-unnatural-instructions-7b

kind
original
quant
bf16 (16-bit)
family
misc
modality
text-generation
root model
allenai/open-instruct-unnatural-instructions-7b
provider
allenai
license
—
language
en
context
—
library
—
vRAM (GB)
—
size on disk
—
downloads
59
likes
1
created
—
last modified
2023-06-20T17:50:10+00:00 (2023)
huggingface
https://huggingface.co/allenai/open-instruct-unnatural-instructions-7b
tags
transformerspytorchllamatext-generationendataset:mrm8488/unnatural-instructionsarxiv:2306.04751arxiv:2302.13971arxiv:2212.09689text-generation-inferenceendpoints_compatibleregion:us

Benchmark scores (4)

benchmarkcategoryscorecontextconfevidence
MMLUknowledge42.9GSM CoT | BBH CoT | TydiQA Closed-booklow| 42.9 | 38.1 | 3.5 | 5.0 | 31.4 | 30.0 | 36.3 | 6.5 | 10.3 | 19.8 | 8.2 | 20.0 |
TydiQAknowledge36.3GSM CoT | BBH CoT | TydiQA Closed-booklow| 42.9 | 38.1 | 3.5 | 5.0 | 31.4 | 30.0 | 36.3 | 6.5 | 10.3 | 19.8 | 8.2 | 20.0 |
BBHreasoning31.4GSM CoT | BBH CoT | TydiQA Closed-booklow| 42.9 | 38.1 | 3.5 | 5.0 | 31.4 | 30.0 | 36.3 | 6.5 | 10.3 | 19.8 | 8.2 | 20.0 |
Codex-Evalcoding10.3GSM CoT | BBH CoT | TydiQA Closed-booklow| 42.9 | 38.1 | 3.5 | 5.0 | 31.4 | 30.0 | 36.3 | 6.5 | 10.3 | 19.8 | 8.2 | 20.0 |

Discussions (0)

none scraped.

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