ChimeraLlama-3-8B-v2

Home/Models/mlabonne/ChimeraLlama-3-8B-v2

mlabonne/ChimeraLlama-3-8B-v2

kind: merge
quant: merge
family: llama
modality: text-generation
root model: NousResearch/Meta-Llama-3-8B-Instruct
provider: mlabonne
license: other
language: —
context: —
library: —
vRAM (GB): —
size on disk: —
downloads: 8,021
likes: 14
created: —
last modified: 2024-09-03T12:41:36+00:00 (2024)
huggingface: https://huggingface.co/mlabonne/ChimeraLlama-3-8B-v2
tags: transformerssafetensorsllamatext-generationmergemergekitlazymergekitbase_model:Locutusque/llama-3-neural-chat-v1-8bbase_model:merge:Locutusque/llama-3-neural-chat-v1-8bbase_model:NousResearch/Meta-Llama-3-8B-Instructbase_model:merge:NousResearch/Meta-Llama-3-8B-Instructbase_model:cloudyu/Meta-Llama-3-8B-Instruct-DPObase_model:merge:cloudyu/Meta-Llama-3-8B-Instruct-DPObase_model:dphn/dolphin-2.9-llama3-8bbase_model:merge:dphn/dolphin-2.9-llama3-8bbase_model:mlabonne/OrpoLlama-3-8Bbase_model:merge:mlabonne/OrpoLlama-3-8Bbase_model:vicgalle/Configurable-Llama-3-8B-v0.3base_model:merge:vicgalle/Configurable-Llama-3-8B-v0.3license:othermodel-indextext-generation-inferenceendpoints_compatibleregion:us

benchmark	category	score	context	conf	evidence
IFEval	instruction-following	44.69	IFEval (0-Shot)	medium	`\| IFEval (0-Shot) \| 44.69 \|`
MMLU-Pro	knowledge	28.54	MMLU-PRO (5-shot)	medium	`\| MMLU-PRO (5-shot) \| 28.54 \|`
BBH	reasoning	28.48	BBH (3-Shot)	medium	`\| BBH (3-Shot) \| 28.48 \|`
MATH	math	8.31	MATH Lvl 5 (4-Shot)	medium	`\| MATH Lvl 5 (4-Shot) \| 8.31 \|`
MUSR	reasoning	5.25	MuSR (0-shot)	medium	`\| MuSR (0-shot) \| 5.25 \|`
GPQA	reasoning	4.7	GPQA (0-shot)	medium	`\| GPQA (0-shot) \| 4.70 \|`

none scraped.