Jackalope 7B By openaccess-ai-collective: Benchmarks, Features and Detailed Analysis. Insights on Jackalope 7B.

Arxiv:2301.13688 Arxiv:2306.02707 Autotrain compatible Conversational Dataset:ldjnr/lesswrong-amplif... Dataset:ldjnr/pure-dove Dataset:ldjnr/verified-camel Dataset:meta-math/metamathqa Dataset:open-orca/openorca Dataset:pygmalionai/pippa Dataset:riddle sense En Endpoints compatible Mistral Pytorch Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/openaccess-ai-collective/jackalope-7b

Jackalope 7B Benchmarks

ARC: 63.4 vs 96.7 (so35)^-34.4%

HellaSwag: 83.29 vs 95.3 (gpt4)^-12.6%

MMLU: 63.5 vs 88.3 (so35)^-28.1%

TruthfulQA: 50.06 vs 59 (gpt4)^-15.2%

WinoGrande: 78.06 vs 87.5 (gpt4)^-10.8%

GSM8K: 28.66 vs 96.4 (so35)^-70.3%

LLME Score: 0.14344

^nn.n% — How the model compares to the reference models: Anthropic Sonnet 3.5 ("so35"), GPT-4o ("gpt4o") or GPT-4 ("gpt4").

What is the LLM Explorer Rank (Score)

Jackalope 7B (openaccess-ai-collective/jackalope-7b)

🌟 Advertise your project 🚀

Jackalope 7B Parameters and Internals

Model Type

text generation

Use Cases

Areas:

research, multimodal applications

Applications:

chatbots, automated reasoning

Primary Use Cases:

multi-turn chat, text generation

Supported Languages

en (English)

Training Details

Data Sources:

Open-Orca/OpenOrca, LDJnr/LessWrong-Amplify-Instruct, LDJnr/Pure-Dove, LDJnr/Verified-Camel, PygmalionAI/PIPPA, meta-math/MetaMathQA, riddle_sense

Methodology:

Fine-tuning on SlimOrca dataset, multi-turn chat improved with OpenChat packing, trained with Axolotl

Training Time:

96 hours with 8x A6000 GPUs

Hardware Used:

8x A6000 GPUs

Input Output

Input Format:

OpenAI's ChatML format with special tokens

Accepted Modalities:

text

Performance Tips:

Use 'tokenize=True' and 'return_tensors="pt"' for optimized performance

Release Notes

Notes:

Release highlights the efficiency and improved multi-turn chat capabilities.

LLM Name	Jackalope 7B
Repository 🤗	https://huggingface.co/openaccess-ai-collective/jackalope-7b
Model Size	7b
Required VRAM	14.4 GB
Updated	2025-09-23
Maintainer	openaccess-ai-collective
Model Type	mistral
Model Files	9.9 GB: 1-of-2 4.5 GB: 2-of-2
Supported Languages	en
Model Architecture	MistralForCausalLM
License	apache-2.0
Context Length	32768
Model Max Length	32768
Transformers Version	4.34.0.dev0
Tokenizer Class	LlamaTokenizer
Vocabulary Size	32002
Torch Data Type	bfloat16

Quantized Models of the Jackalope 7B

Model	Likes	Downloads	VRAM
Jackalope 7B GGUF	7	328	3 GB
Jackalope 7B AWQ	1	6	4 GB
Jackalope 7B GPTQ	0	6	4 GB

Best Alternatives to Jackalope 7B

Best Alternatives	Context / RAM	Downloads	Likes
...Nemo Instruct 2407 Abliterated	1000K / 24.5 GB	141	18
MegaBeam Mistral 7B 512K	512K / 14.4 GB	8908	50
SpydazWeb AI HumanAI RP	512K / 14.4 GB	17	1
SpydazWeb AI HumanAI 002	512K / 14.4 GB	18	1
...daz Web AI ChatML 512K Project	512K / 14.5 GB	12	0
MegaBeam Mistral 7B 300K	282K / 14.4 GB	3779	16
MegaBeam Mistral 7B 300K	282K / 14.4 GB	8082	16
Hebrew Mistral 7B 200K	256K / 30 GB	1316	15
Astral 256K 7B V2	250K / 14.4 GB	5	0
Astral 256K 7B	250K / 14.4 GB	5	0

Note: green Score (e.g. "73.2") means that the model is better than openaccess-ai-collective/jackalope-7b.

Rank the Jackalope 7B Capabilities

🆘 Have you tried this model? Rate its performance. This feedback would greatly assist ML community in identifying the most suitable model for their needs. Your contribution really does make a difference! 🌟

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

What open-source LLMs or SLMs are you in search of? 51539 in total.

Email us: info@extractum.io. Our Privacy Policy | Terms and Conditions | Suggest an improvement.

Our Social Media →

Original data from HuggingFace, OpenCompass and various public git repos.

Release v20241124

Support LLM Explorer