phi-3-mini-LoRA / README.md
edumunozsala's picture
Update README.md
779c193 verified
metadata
license: mit
library_name: peft
tags:
  - trl
  - sft
  - generated_from_trainer
  - code
  - python
base_model: microsoft/Phi-3-mini-4k-instruct
model-index:
  - name: phi-3-mini-LoRA
    results: []
datasets:
  - iamtarun/python_code_instructions_18k_alpaca
pipeline_tag: text-generation

Phi-3-mini 3.8B Python Coder 👩‍💻

Phi-3-mini 3.8B fine-tuned on the python_code_instructions_18k_alpaca Code instructions dataset by using the method LoRA with PEFT library.

Pretrained description

Llama-2

The Phi-3-Mini-4K-Instruct is a 3.8B parameters, lightweight, state-of-the-art open model trained with the Phi-3 datasets that includes both synthetic data and the filtered publicly available websites data with a focus on high-quality and reasoning dense properties. The model belongs to the Phi-3 family with the Mini version in two variants 4K and 128K which is the context length (in tokens) that it can support.

Tokenizer

Phi-3 Mini-4K-Instruct supports a vocabulary size of up to 32064 tokens. The tokenizer files already provide placeholder tokens that can be used for downstream fine-tuning, but they can also be extended up to the model's vocabulary size.

Training data

python_code_instructions_18k_alpaca

The dataset contains problem descriptions and code in python language. This dataset is taken from sahil2801/code_instructions_120k, which adds a prompt column in alpaca style.

Chat Format

Given the nature of the training data, the Phi-3 Mini-4K-Instruct model is best suited for prompts using the chat format as follows. You can provide the prompt as a question with a generic template as follow:

<|user|>\nQuestion <|end|>\n<|assistant|>

For example:

<|user|>
How to explain Internet for a medieval knight?<|end|>
<|assistant|>

where the model generates the text after <|assistant|> . In case of few-shots prompt, the prompt can be formatted as the following:

<|user|>
I am going to Paris, what should I see?<|end|>
<|assistant|>
Paris, the capital of France, is known for its stunning architecture, art museums, historical landmarks, and romantic atmosphere. Here are some of the top attractions to see in Paris:\n\n1. The Eiffel Tower: The iconic Eiffel Tower is one of the most recognizable landmarks in the world and offers breathtaking views of the city.\n2. The Louvre Museum: The Louvre is one of the world's largest and most famous museums, housing an impressive collection of art and artifacts, including the Mona Lisa.\n3. Notre-Dame Cathedral: This beautiful cathedral is one of the most famous landmarks in Paris and is known for its Gothic architecture and stunning stained glass windows.\n\nThese are just a few of the many attractions that Paris has to offer. With so much to see and do, it's no wonder that Paris is one of the most popular tourist destinations in the world."<|end|>
<|user|>
What is so great about #1?<|end|>
<|assistant|>

Training hyperparameters

The following hyperparameters were used during training:

  • learning_rate: 0.0001
  • train_batch_size: 8
  • eval_batch_size: 8
  • seed: 42
  • gradient_accumulation_steps: 4
  • total_train_batch_size: 32
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: linear
  • lr_scheduler_warmup_ratio: 0.1
  • num_epochs: 3

Training results

Training Loss Epoch Step Validation Loss
1.1716 0.1809 100 0.6639
0.6253 0.3618 200 0.5865
0.5772 0.5427 300 0.5753
0.5823 0.7237 400 0.5703
0.5862 0.9046 500 0.5673
0.5804 1.0855 600 0.5652
0.5776 1.2664 700 0.5641
0.5721 1.4473 800 0.5630
0.5725 1.6282 900 0.5623
0.5708 1.8091 1000 0.5615
0.5714 1.9900 1100 0.5611
0.5685 2.1710 1200 0.5607
0.5618 2.3519 1300 0.5605
0.5789 2.5328 1400 0.5605
0.5716 2.7137 1500 0.5600
0.5626 2.8946 1600 0.5601

Framework versions

  • PEFT 0.10.0
  • Transformers 4.40.1
  • Pytorch 2.2.1+cu121
  • Datasets 2.19.0
  • Tokenizers 0.19.1