Stablelm Tuned Alpha 3B By stabilityai: Benchmarks, Features and Detailed Analysis. Insights on Stablelm Tuned Alpha 3B.

Autotrain compatible Dataset:dahoas/full-hh-rlhf Dataset:dmayhem93/chatcombined Dataset:huggingfaceh4/databric... Dataset:jeffwan/sharegpt vicun... Dataset:nomic-ai/gpt4all promp... Dataset:tatsu-lab/alpaca En Endpoints compatible Gpt neox Pytorch Region:us Sharded

Model Card on HF 🤗: https://huggingface.co/stabilityai/stablelm-tuned-alpha-3b

Stablelm Tuned Alpha 3B Benchmarks

ARC: 27.82 vs 96.7 (so35)^-71.2%

HellaSwag: 44.06 vs 95.3 (gpt4)^-53.8%

MMLU: 23.08 vs 88.3 (so35)^-73.9%

TruthfulQA: 42.33 vs 59 (gpt4)^-28.3%

WinoGrande: 55.01 vs 87.5 (gpt4)^-37.1%

GSM8K: 0.53 vs 96.4 (so35)^-99.5%

LLME Score: 0.15006

^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)

Stablelm Tuned Alpha 3B (stabilityai/stablelm-tuned-alpha-3b)

🌟 Advertise your project 🚀

Stablelm Tuned Alpha 3B Parameters and Internals

Model Type

causal-lm

Use Cases

Areas:

open-source community, chat-like applications

Limitations:

The model may generate biased or toxic text despite efforts in safe fine-tuning., Not intended as a replacement for human judgment

Considerations:

Be mindful of potential bias or toxic outputs.

Additional Notes

Models include a helpful hand from Dakota Mahan ([@dmayhem93](https://huggingface.co/dmayhem93)) in their development.

Supported Languages

English (Proficient)

Training Details

Data Sources:

tatsu-lab/alpaca, nomic-ai/gpt4all_prompt_generations, Dahoas/full-hh-rlhf, jeffwan/sharegpt_vicuna, HuggingFaceH4/databricks_dolly_15k

Methodology:

Supervised fine-tuning on natural language datasets focused on chat and instruction-following tasks.

Context Length:

4096

Model Architecture:

NeoX transformer architecture

Responsible Ai Considerations

Fairness:

Models are developed to adhere to safer distributions of text but cannot mitigate all biases and toxicity.

Transparency:

It should not be treated as a substitute for human judgment or considered a source of truth.

Accountability:

Users are responsible for the outputs generated and should use models responsibly.

Mitigation Strategies:

Fine-tuning on datasets aimed at improving safety, but may not remove all biases/toxicity.

Input Output

Input Format:

Prompts formatted to <|SYSTEM|>...<|USER|>...<|ASSISTANT|>...

Accepted Modalities:

text

Output Format:

Text output

LLM Name	Stablelm Tuned Alpha 3B
Repository 🤗	https://huggingface.co/stabilityai/stablelm-tuned-alpha-3b
Model Size	3b
Required VRAM	14.9 GB
Updated	2025-08-17
Maintainer	stabilityai
Model Type	gpt_neox
Model Files	10.2 GB: 1-of-2 4.7 GB: 2-of-2
Supported Languages	en
Model Architecture	GPTNeoXForCausalLM
License	cc-by-nc-sa-4.0
Context Length	4096
Model Max Length	4096
Transformers Version	4.28.1
Tokenizer Class	GPTNeoXTokenizer
Vocabulary Size	50688
Torch Data Type	float32

Quantized Models of the Stablelm Tuned Alpha 3B

Model	Likes	Downloads	VRAM
Stablelm Tuned Alpha 3B 8bit	3	14	4 GB
Stablelm Tuned Alpha 3B 16bit	6	19	7 GB

Best Alternatives to Stablelm Tuned Alpha 3B

Best Alternatives	Context / RAM	Downloads	Likes
Stablecode Completion Alpha 3B	16K / 14.1 GB	25	118
RedPajama 3B 16384	16K / 19.7 GB	16	5
Redpajama 3B Chat	5K / 6.4 GB	32	2
Stablelm Base Alpha 3B	4K / 14.9 GB	3200	82
...blecode Completion Alpha 3B 4K	4K / 6.1 GB	2044	280
Stablecode Instruct Alpha 3B	4K / 6.1 GB	5	304
StableCode 3B	4K / 6.1 GB	17	1
...tion Alpha 3B 4K Openvino Int8	4K / 2.8 GB	12	1
Redpajama 3B Evol Coder	4K / 6.1 GB	15	1
Literature 3B 4096	4K / 11.7 GB	17	7

Note: green Score (e.g. "73.2") means that the model is better than stabilityai/stablelm-tuned-alpha-3b.

Rank the Stablelm Tuned Alpha 3B 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? 50728 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