DesiLadkaa/indian-finance-stage2-sft-merged
DesiLadkaa/indian-finance-stage2-sft-merged is a 1.5 billion parameter Qwen2 model developed by DesiLadkaa, fine-tuned from DesiLadkaa/indian-finance-stage1-noninstruct-merged. This model is specifically optimized for financial applications within the Indian context, leveraging a 32768 token context length. It was trained using Unsloth and Huggingface's TRL library, enabling faster fine-tuning.
Loading preview...
Overview
DesiLadkaa/indian-finance-stage2-sft-merged is a 1.5 billion parameter Qwen2 model, developed by DesiLadkaa. It is a fine-tuned version of the DesiLadkaa/indian-finance-stage1-noninstruct-merged model, specifically designed for applications in Indian finance.
Key Characteristics
- Architecture: Qwen2 model family.
- Parameter Count: 1.5 billion parameters.
- Context Length: Supports a substantial context window of 32768 tokens.
- Training Efficiency: Fine-tuned using Unsloth and Huggingface's TRL library, which facilitated a 2x faster training process.
- License: Released under the Apache-2.0 license.
Primary Use Case
This model is tailored for specialized tasks within the Indian financial domain, building upon its stage 1 non-instruct predecessor. Its fine-tuned nature suggests enhanced performance for specific financial language understanding and generation tasks relevant to the Indian market.