HF Model & Benchmark Matrix

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

CarbonVillain-en-10.7B-v4-GGUF

TheBloke/CarbonVillain-en-10.7B-v4-GGUF

kind
gguf
quant
gguf (2-bit)
family
misc
modality
—
root model
jeonsworld/CarbonVillain-en-10.7B-v4
provider
TheBloke
license
mit
language
en
context
4096,
library
—
vRAM (GB)
—
size on disk
—
downloads
658
likes
6
created
—
last modified
2023-12-31T20:16:53+00:00 (2023)
huggingface
https://huggingface.co/TheBloke/CarbonVillain-en-10.7B-v4-GGUF
tags
transformersggufsolarenbase_model:jeonsworld/CarbonVillain-en-10.7B-v4base_model:quantized:jeonsworld/CarbonVillain-en-10.7B-v4license:mitregion:usconversational

Benchmark scores (5)

benchmarkcategoryscorecontextconfevidence
HellaSwagknowledge88.48HellaSwag (10-shot)medium| HellaSwag (10-shot) | 88.48 |
WinoGrandeknowledge83.58Winogrande (5-shot)medium| Winogrande (5-shot) | 83.58 |
TruthfulQAsafety71.95TruthfulQA (0-shot)medium| TruthfulQA (0-shot) | 71.95 |
MMLUknowledge66.27MMLU (5-shot)medium| MMLU (5-shot) | 66.27 |
GSM8Kmath65.58GSM8K (5-shot)medium| GSM8K (5-shot) | 65.58 |

Discussions (0)

none scraped.

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