HF Model & Benchmark Matrix

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

gemma-2b

google/gemma-2b

kind
gguf
quant
gguf
family
gemma
modality
text-generation
root model
google/gemma-2b-it-pytorch
provider
google
license
gemma
language
—
context
8192
library
transformers
vRAM (GB)
—
size on disk
—
downloads
97,106
likes
1,209
created
—
last modified
2024-09-27T12:18:55+00:00 (2024)
huggingface
https://huggingface.co/google/gemma-2b
tags
transformerssafetensorsggufgemmatext-generationarxiv:2312.11805arxiv:2009.03300arxiv:1905.07830arxiv:1911.11641arxiv:1904.09728arxiv:1905.10044arxiv:1907.10641arxiv:1811.00937arxiv:1809.02789arxiv:1911.01547arxiv:1705.03551arxiv:2107.03374arxiv:2108.07732arxiv:2110.14168arxiv:2304.06364arxiv:2206.04615arxiv:1804.06876arxiv:2110.08193arxiv:2009.11462

Benchmark scores (14)

benchmarkcategoryscorecontextconfevidence
PIQAknowledge77.3[PIQA](https://arxiv.org/abs/1911.11641)medium| [PIQA](https://arxiv.org/abs/1911.11641) | 0-shot | 77.3 | 81.2 |
HellaSwagknowledge71.4[HellaSwag](https://arxiv.org/abs/1905.07830)medium| [HellaSwag](https://arxiv.org/abs/1905.07830) | 0-shot | 71.4 | 81.2 |
WinoGrandeknowledge65.4[WinoGrande](https://arxiv.org/abs/1907.10641)medium| [WinoGrande](https://arxiv.org/abs/1907.10641) | partial score | 65.4 | 72.3 |
CommonsenseQAknowledge65.3[CommonsenseQA](https://arxiv.org/abs/1811.00937)medium| [CommonsenseQA](https://arxiv.org/abs/1811.00937) | 7-shot | 65.3 | 71.3 |
TriviaQAknowledge53.2[TriviaQA](https://arxiv.org/abs/1705.03551)medium| [TriviaQA](https://arxiv.org/abs/1705.03551) | 5-shot | 53.2 | 63.4 |
OpenBookQAknowledge47.8[OpenBookQA](https://arxiv.org/abs/1809.02789)medium| [OpenBookQA](https://arxiv.org/abs/1809.02789) | | 47.8 | 52.8 |
MMLUknowledge42.3[MMLU](https://arxiv.org/abs/2009.03300)medium| [MMLU](https://arxiv.org/abs/2009.03300) | 5-shot, top-1 | 42.3 | 64.3 |
BigBenchreasoning35.2[BIG-Bench](https://arxiv.org/abs/2206.04615)medium| [BIG-Bench](https://arxiv.org/abs/2206.04615) | | 35.2 | 55.1 |
TruthfulQAsafety31.81[TruthfulQA](https://arxiv.org/abs/2109.07958)medium| [TruthfulQA](https://arxiv.org/abs/2109.07958) | | 31.81 | 44.84 |
MBPPcoding29.2[MBPP](https://arxiv.org/abs/2108.07732)medium| [MBPP](https://arxiv.org/abs/2108.07732) | 3-shot | 29.2 | 44.4 |
AGIEvalreasoning24.2[AGIEval](https://arxiv.org/abs/2304.06364)medium| [AGIEval](https://arxiv.org/abs/2304.06364) | | 24.2 | 41.7 |
HumanEvalcoding22.0[HumanEval](https://arxiv.org/abs/2107.03374)medium| [HumanEval](https://arxiv.org/abs/2107.03374) | pass@1 | 22.0 | 32.3 |
GSM8Kmath17.7[GSM8K](https://arxiv.org/abs/2110.14168)medium| [GSM8K](https://arxiv.org/abs/2110.14168) | maj@1 | 17.7 | 46.4 |
MATHmath11.8[MATH](https://arxiv.org/abs/2108.07732)medium| [MATH](https://arxiv.org/abs/2108.07732) | 4-shot | 11.8 | 24.3 |

Discussions (0)

none scraped.

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