Nidum-Llama-3.2-3B-Uncensored

Welcome to Nidum!

At Nidum, we believe in pushing the boundaries of innovation by providing advanced and unrestricted AI models for every application. Dive into our world of possibilities and experience the freedom of Nidum-Llama-3.2-3B-Uncensored, tailored to meet diverse needs with exceptional performance.


GitHub Icon
Explore Nidum's Open-Source Projects on GitHub: https://github.com/NidumAI-Inc


Key Features

  1. Uncensored Responses: Capable of addressing any query without content restrictions, offering detailed and uninhibited answers.
  2. Versatility: Excels in diverse use cases, from complex technical queries to engaging casual conversations.
  3. Advanced Contextual Understanding: Draws from an expansive knowledge base for accurate and context-aware outputs.
  4. Extended Context Handling: Optimized for handling long-context interactions for improved continuity and depth.
  5. Customizability: Adaptable to specific tasks and user preferences through fine-tuning.

Use Cases

  • Open-Ended Q&A
  • Creative Writing and Ideation
  • Research Assistance
  • Educational Queries
  • Casual Conversations
  • Mathematical Problem Solving
  • Long-Context Dialogues

How to Use

To start using Nidum-Llama-3.2-3B-Uncensored, follow the sample code below:

import torch
from transformers import pipeline

pipe = pipeline(
    "text-generation",
    model="nidum/Nidum-Llama-3.2-3B-Uncensored",
    model_kwargs={"torch_dtype": torch.bfloat16},
    device="cuda",  # replace with "mps" to run on a Mac device
)

messages = [
    {"role": "user", "content": "Tell me something fascinating."},
]

outputs = pipe(messages, max_new_tokens=256)
assistant_response = outputs[0]["generated_text"][-1]["content"].strip()
print(assistant_response)

Quantized Models Available for Download

Quantized Model Version Description
Nidum-Llama-3.2-3B-Uncensored-F16.gguf Full 16-bit floating point precision for maximum accuracy on high-end GPUs.
model-Q2_K.gguf Optimized for minimal memory usage with lower precision, suitable for edge cases.
model-Q3_K_L.gguf Balanced precision with enhanced memory efficiency for medium-range devices.
model-Q3_K_M.gguf Mid-range quantization for moderate precision and memory usage balance.
model-Q3_K_S.gguf Smaller quantization steps, offering moderate precision with reduced memory use.
model-Q4_0_4_4.gguf Performance-optimized for low memory, ideal for lightweight deployment.
model-Q4_0_4_8.gguf Extended quantization balancing memory use and inference speed.
model-Q4_0_8_8.gguf Advanced memory precision targeting larger contexts.
model-Q4_K_M.gguf High-efficiency quantization for moderate GPU resources.
model-Q4_K_S.gguf Optimized for smaller-scale operations with compact memory footprint.
model-Q5_K_M.gguf Balances performance and precision, ideal for robust inferencing environments.
model-Q5_K_S.gguf Moderate quantization targeting performance with minimal resource usage.
model-Q6_K.gguf High-precision quantization for accurate and stable inferencing tasks.
model-TQ1_0.gguf Experimental quantization for targeted applications in test environments.
model-TQ2_0.gguf High-performance tuning for experimental use cases and flexible precision.

Datasets and Fine-Tuning

The following fine-tuning datasets are leveraged to enhance specific model capabilities:

  • Uncensored Data: Enables unrestricted and uninhibited responses.
  • RAG-Based Fine-Tuning: Optimizes retrieval-augmented generation for knowledge-intensive tasks.
  • Long Context Fine-Tuning: Enhances the model's ability to process and maintain coherence in extended conversations.
  • Math-Instruct Data: Specially curated for precise and contextually accurate mathematical reasoning.

Benchmarks

After fine-tuning with uncensored data, Nidum-Llama-3.2-3B demonstrates superior performance compared to the original LLaMA model, particularly in accuracy and handling diverse, unrestricted scenarios.

Benchmark Summary Table

Benchmark Metric LLaMA 3.2 3B Nidum 3.2 3B Observation
GPQA Exact Match (Flexible) 0.3 0.5 Nidum 3B demonstrates significant improvement, particularly in generative tasks.
Accuracy 0.4 0.5 Consistent improvement, especially in zero-shot scenarios.
HellaSwag Accuracy 0.3 0.4 Better performance in common sense reasoning tasks.
Normalized Accuracy 0.3 0.4 Enhanced ability to understand and predict context in sentence completion.
Normalized Accuracy (Stderr) 0.15275 0.1633 Slightly improved consistency in normalized accuracy.
Accuracy (Stderr) 0.15275 0.1633 Shows robustness in reasoning accuracy compared to LLaMA 3B.

Insights:

  1. GPQA Results: Fine-tuning on uncensored data has boosted Nidum 3B's Exact Match and Accuracy, particularly excelling in generative and zero-shot tasks involving domain-specific knowledge.
  2. HellaSwag Results: Nidum 3B consistently outperforms LLaMA 3B in common sense reasoning benchmarks, indicating enhanced contextual and semantic understanding.

Contributing

We welcome contributions to improve and extend the model’s capabilities. Stay tuned for updates on how to contribute.


Contact

For inquiries, collaborations, or further information, please reach out to us at info@nidum.ai.


Explore the Possibilities

Dive into unrestricted creativity and innovation with Nidum Llama 3.2 3B Uncensored!

Downloads last month
2,077
GGUF
Model size
3.21B params
Architecture
llama

2-bit

3-bit

4-bit

5-bit

6-bit

16-bit

Inference Examples
Inference API (serverless) does not yet support adapter-transformers models for this pipeline type.

Model tree for nidum/Nidum-Llama-3.2-3B-Uncensored-GGUF

Adapter
(42)
this model

Collection including nidum/Nidum-Llama-3.2-3B-Uncensored-GGUF