donoway/BoolQ_Llama-3.2-1B-26t8ytsb
The donoway/BoolQ_Llama-3.2-1B-26t8ytsb model is a 1 billion parameter language model fine-tuned from Meta Llama-3.2-1B. It achieved an accuracy of 0.7147 on its evaluation set, demonstrating capabilities in tasks likely related to boolean question answering. This model is suitable for applications requiring a compact, specialized language model with a 32768 token context length.
Loading preview...
Model Overview
The donoway/BoolQ_Llama-3.2-1B-26t8ytsb is a 1 billion parameter language model, fine-tuned from the meta-llama/Llama-3.2-1B architecture. It was trained for 100 epochs with a learning rate of 2e-05 and a batch size of 32. The model demonstrates an overall accuracy of 0.7147 on its evaluation set, with a generated accuracy of 0.7040.
Key Performance Metrics
- Overall Accuracy: 0.7147
- Generated Accuracy: 0.7040
- Validation Loss: 1.6420 (at best epoch)
Training Details
The model was trained using the AdamW optimizer with cosine learning rate scheduling and a warmup ratio of 0.01. The training process involved 100 epochs, with evaluation metrics logged at various steps, showing fluctuations in accuracy and loss across different subsets of the evaluation data.
Intended Use Cases
While specific intended uses are not detailed in the README, the model's name and evaluation metrics suggest its potential for tasks involving boolean question answering or similar classification-based language understanding. Its compact size (1B parameters) makes it suitable for deployment in resource-constrained environments or applications where a smaller, specialized model is preferred over larger, general-purpose LLMs.