HF Model & Benchmark Matrix

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

Synatra-7B-v0.3-dpo-GGUF

TheBloke/Synatra-7B-v0.3-dpo-GGUF

kind
gguf
quant
gguf (2-bit)
family
misc
modality
—
root model
maywell/Synatra-7B-v0.3-dpo
provider
TheBloke
license
cc-by-sa-4.0
language
—
context
—
library
—
vRAM (GB)
—
size on disk
—
downloads
140
likes
5
created
—
last modified
2023-11-23T16:17:42+00:00 (2023)
huggingface
https://huggingface.co/TheBloke/Synatra-7B-v0.3-dpo-GGUF
tags
transformersggufmistralbase_model:maywell/Synatra-7B-v0.3-dpobase_model:quantized:maywell/Synatra-7B-v0.3-dpolicense:cc-by-sa-4.0region:usconversational

Benchmark scores (6)

benchmarkcategoryscorecontextconfevidence
HellaSwagknowledge82.58HellaSwag (10-shot)medium| HellaSwag (10-shot) | 82.58 |
WinoGrandeknowledge76.24Winogrande (5-shot)medium| Winogrande (5-shot) | 76.24 |
MMLUknowledge61.46MMLU (5-shot)medium| MMLU (5-shot) | 61.46 |
TruthfulQAsafety56.46TruthfulQA (0-shot)medium| TruthfulQA (0-shot) | 56.46 |
GSM8Kmath23.73GSM8K (5-shot)medium| GSM8K (5-shot) | 23.73 |
DROPreasoning8.68DROP (3-shot)medium| DROP (3-shot) | 8.68 |

Discussions (0)

none scraped.

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