Model Card for gemma-270-it-vcet-lora

Model Details

Model Description

This model is a domain-specific conversational AI fine-tuned on custom data related to VCET College, Madurai. Built on top of unsloth/gemma-3-270m-it-unsloth-bnb-4bit, it uses LoRA and PEFT for efficient adaptation. The model is designed to answer queries about campus life, academics, departments, events, and administrative processes at VCET.

• Developed by: SandeepCodez.
• Funded by [optional]: Self-funded.
• Shared by [optional]: SandeepCodez
• Model type: Causal Language Model (Text Generation).
• Language(s) (NLP): English (with contextual Tamil understanding).
• License: Apache 2.0.
• Finetuned from model [optional]: unsloth/gemma-3-270m-it-unsloth-bnb-4bit

Model Sources [optional]

• Repository: https://github.com/SandeepCodez
• Paper [optional]: [More Information Needed]
• Demo [optional]: [More Information Needed]

Uses

Direct Use

Answering VCET-related questions

Assisting students with academic and campus queries

Automating college FAQs

Supporting chatbot integration for VCET platforms

[More Information Needed]

Downstream Use [optional]

Integration into college ERP systems

Enhancing virtual assistants for student support

Embedding in mobile apps or websites

[More Information Needed]

Out-of-Scope Use

General-purpose text generation outside VCET context

Legal, medical, or financial advice

High-stakes decision-making without human oversight

[More Information Needed]

Bias, Risks, and Limitations

May reflect institutional bias from VCET sources

Limited generalization outside VCET domain

Not suitable for sensitive or critical applications

[More Information Needed]

Recommendations

Use in supervised environments

Periodic updates to dataset recommended

Human validation for factual accuracy advised

Users (both direct and downstream) should be made aware of the risks, biases and limitations of the model. More information needed for further recommendations.

How to Get Started with the Model

from transformers import AutoModelForCausalLM, AutoTokenizer

model_id = "SandeepCodez/gemma-270-it-vcet-lora" tokenizer = AutoTokenizer.from_pretrained(model_id) model = AutoModelForCausalLM.from_pretrained(model_id)

inputs = tokenizer("What are the placement statistics for VCET Madurai?", return_tensors="pt") outputs = model.generate(**inputs) print(tokenizer.decode(outputs[0]))

Training Details

Training Data

Custom dataset created by the developer, including:

VCET brochures

Departmental documents

Student interviews

Campus FAQs

Event archives

[More Information Needed]

Training Procedure

Preprocessing [optional]

Cleaned and structured into JSONL format

Tokenized using Gemma tokenizer

Filtered for relevance and clarity

Training Hyperparameters

Training regime: bf16 mixed precision

Epochs: 3

Batch Size: 16

Learning Rate: 2e-4

Frameworks: PEFT 0.17.1, TRL, Unsloth

Speeds, Sizes, Times [optional]

Training Time: ~3 hours

Dataset Size: ~10,000 samples

Model Size: 270M parameters

[More Information Needed]

Evaluation

Testing Data, Factors & Metrics

Testing Data

[More Information Needed]

Factors

[More Information Needed]

Metrics

[More Information Needed]

Results

[More Information Needed]

Summary

Model Examination [optional]

[More Information Needed]

Environmental Impact

Carbon emissions can be estimated using the Machine Learning Impact calculator presented in Lacoste et al. (2019).

• Hardware Type: [More Information Needed]
• Hours used: [More Information Needed]
• Cloud Provider: [More Information Needed]
• Compute Region: [More Information Needed]
• Carbon Emitted: [More Information Needed]

Technical Specifications [optional]

Model Architecture and Objective

[More Information Needed]

Compute Infrastructure

[More Information Needed]

Hardware

[More Information Needed]

Software

[More Information Needed]

Citation [optional]

BibTeX:

[More Information Needed]

APA:

[More Information Needed]

Glossary [optional]

[More Information Needed]

More Information [optional]

[More Information Needed]

Model Card Authors [optional]

[More Information Needed]

Model Card Contact

[More Information Needed]

Framework versions

• PEFT 0.17.1