Falcon Think3 10 B I T

Introduction

FalconThink3-10B-IT is a specialized model fine-tuned on the Ultrathink dataset. This model is a demonstration of using this dataset for text generation and reasoning tasks. It builds upon the Falcon3-10B-Instruct base model.

Architecture

• Base Model: tiiuae/Falcon3-10B-Instruct
• Model Type: LlamaForCausalLM
• Parameters: 10.3 billion
• Precision: FP16
• Languages: Primarily English

Training

The model was trained using the SFT (Supervised Fine-Tuning) method on the fluently-sets/ultrathink dataset. The training was conducted by the Fluently Team in collaboration with Unsloth AI. The model is designed to enhance text generation capabilities and reasoning.

Guide: Running Locally

1. Setup Environment: Ensure you have Python and necessary libraries installed. Use virtual environments for managing dependencies.

2. Download the Model: Use Hugging Face's transformers library to download the model.

   from transformers import AutoModelForCausalLM, AutoTokenizer
   
   model_name = "fluently-sets/FalconThink3-10B-IT"
   model = AutoModelForCausalLM.from_pretrained(model_name)
   tokenizer = AutoTokenizer.from_pretrained(model_name)
   

3. Run Inference: Prepare your input text and perform inference with the model.

   input_text = "Your input text here"
   inputs = tokenizer(input_text, return_tensors="pt")
   outputs = model.generate(**inputs)
   print(tokenizer.decode(outputs[0]))
   

4. Hardware Recommendations: For optimal performance, using cloud GPUs such as those provided by AWS, Google Cloud, or Azure is recommended. A GPU with at least 16GB of memory is advisable for handling this model.

License

This model is distributed under the Apache 2.0 License, allowing for both academic and commercial use with proper attribution.