README.md

metadata

license: apache-2.0
tags:
  - phi-2
  - lora
  - merged-model
  - environmental-factors
  - causal-lm
  - transformer

Phi-2 Merged Model: Environmental Emission Factors

This model is a merged version of Microsoft's Phi-2 fine-tuned using LoRA adapters on a custom instruction dataset focused on environmental emission factors.

📌 Use Cases

• Query emission factors for different energy sources by country or region
• Develop mobile or web-based environmental awareness tools
• Educational and research applications around climate data

🧠 Model Architecture

• Base Model: microsoft/phi-2
• Fine-tuning: LoRA (Low-Rank Adaptation)
• Merged: LoRA weights merged into base model using PeftModel.merge_and_unload()

🚀 Usage

Example Python code to generate responses:

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("Surendra-Aitest/phi2-env-factors-merged")

tokenizer = AutoTokenizer.from_pretrained("Surendra-Aitest/phi2-env-factors-merged")

prompt = "What is the CO2 emission factor for biomass energy in Germany?"

inputs = tokenizer(prompt, return_tensors="pt")

outputs = model.generate(**inputs, max_new_tokens=100)

print(tokenizer.decode(outputs[0]))

📂 Files Included

• model-00001-of-00002.safetensors and model-00002-of-00002.safetensors
• tokenizer.json, tokenizer_config.json, vocab.json, merges.txt
• generation_config.json, config.json

📬 Contact

For questions, contact: Surendra-Aitest