Fine-tuned Llama 3.1 8B Instruct Model

This is a fine-tuned version of Meta's Llama-3.1-8B-Instruct model using LoRA (Low-Rank Adaptation) technique.

Model Details

Model Description

This model is a fine-tuned version of Llama 3.1 8B Instruct using PEFT (Parameter-Efficient Fine-Tuning) with LoRA. The fine-tuning process focused on adapting the base model to specific use cases while maintaining the general capabilities of the original model.

  • Developed by: Bhavya Bipin Gada (PRESHENT)
  • Model type: LoRA fine-tuned Llama 3.1
  • Language(s): English
  • License: MIT
  • Finetuned from model: meta-llama/Llama-3.1-8B-Instruct

Training Details

Training Procedure

The model was fine-tuned using the following LoRA configuration:

  • LoRA rank (r): 8
  • LoRA alpha: 16
  • LoRA dropout: 0.05
  • Target modules: k_proj, q_proj, v_proj, o_proj, gate_proj, up_proj, down_proj
  • Task type: CAUSAL_LM

Training Data


[
  {
    "messages": [
      {
        "role": "user",
        "content": "What sustainable solutions do you offer for solar energy?"
      },
      {
        "role": "assistant",
        "content": "We offer rooftop solar panels, community solar options, and solar battery storage solutions optimized for your location."
      }
    ]
  },
  {
    "messages": [
      {
        "role": "user",
        "content": "Can you recommend a sustainable water purification method?"
      },
      {
        "role": "assistant",
        "content": "Reverse osmosis systems powered by renewable energy are highly effective and sustainable."
      }
    ]
  }
]

Usage

You can use this model with the Transformers library:

from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel, PeftConfig

# Load the base model
base_model_id = "meta-llama/Llama-3.1-8B-Instruct"
model = AutoModelForCausalLM.from_pretrained(base_model_id)
tokenizer = AutoTokenizer.from_pretrained(base_model_id)

# Load the LoRA adapter
adapter_path = "bhavyabgada/preshent-llama"
model = PeftModel.from_pretrained(model, adapter_path)

# Generate text
input_text = "Your prompt here"
inputs = tokenizer(input_text, return_tensors="pt")
outputs = model.generate(**inputs, max_length=100)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Limitations and Biases

This model inherits the limitations and biases from the base Llama 3.1 8B Instruct model.

Framework versions

  • PEFT 0.13.2
Downloads last month
10
Inference Providers NEW
This model is not currently available via any of the supported Inference Providers.
The model cannot be deployed to the HF Inference API: The model has no pipeline_tag.

Model tree for bhavyabgada/preshent-llama

Adapter
(809)
this model