Daredevil-7B

Home/Models/mlabonne/Daredevil-7B

mlabonne/Daredevil-7B

kind: merge
quant: merge
family: misc
modality: text-generation
root model: mistralai/Mistral-7B-v0.1
provider: mlabonne
license: cc-by-nc-4.0
language: —
context: —
library: —
vRAM (GB): —
size on disk: —
downloads: 123
likes: 12
created: —
last modified: 2024-03-04T15:16:44+00:00 (2024)
huggingface: https://huggingface.co/mlabonne/Daredevil-7B
tags: transformerssafetensorsmistraltext-generationmergemergekitlazymergekitsamir-fama/SamirGPT-v1abacusai/Slerp-CM-mist-dpoEmbeddedLLM/Mistral-7B-Merge-14-v0.2base_model:EmbeddedLLM/Mistral-7B-Merge-14-v0.2base_model:merge:EmbeddedLLM/Mistral-7B-Merge-14-v0.2base_model:abacusai/Slerp-CM-mist-dpobase_model:merge:abacusai/Slerp-CM-mist-dpobase_model:mistralai/Mistral-7B-v0.1base_model:merge:mistralai/Mistral-7B-v0.1base_model:samir-fama/SamirGPT-v1base_model:merge:samir-fama/SamirGPT-v1license:cc-by-nc-4.0model-indextext-generation-inferenceendpoints_compatibleregion:us

benchmark	category	score	context	conf	evidence
HellaSwag	knowledge	87.17	HellaSwag (10-Shot)	medium	`\| HellaSwag (10-Shot) \| 87.17 \|`
WinoGrande	knowledge	81.29	Winogrande (5-shot)	medium	`\| Winogrande (5-shot) \| 81.29 \|`
GSM8K	math	72.93	GSM8k (5-shot)	medium	`\| GSM8k (5-shot) \| 72.93 \|`
MMLU	knowledge	65.3	MMLU (5-Shot)	medium	`\| MMLU (5-Shot) \| 65.30 \|`
TruthfulQA	safety	64.09	TruthfulQA (0-shot)	medium	`\| TruthfulQA (0-shot) \| 64.09 \|`
TruthfulQA	safety	52.99		low	`\| [OpenHermes-2.5-Mistral-7B](https://huggingface.co/teknium/OpenHermes-2.5-Mistral-7B) \| 42.75 \| 72.99 \| 52.99 \| 40.94 \| 52.42 \|`
BigBench	reasoning	47.07		low	`\| [Daredevil-7B](https://huggingface.co/shadowml/Daredevil-7B) \| 44.85 \| 76.07 \| <u>64.89</u> \| 47.07 \| <u>58.22</u> \|`
AGIEval	reasoning	44.85		low	`\| [Daredevil-7B](https://huggingface.co/shadowml/Daredevil-7B) \| 44.85 \| 76.07 \| <u>64.89</u> \| 47.07 \| <u>58.22</u> \|`

none scraped.