README.md

---
library_name: peft
base_model: meta-llama/Llama-2-7b-hf
license: mit
language:
- en
metrics:
- bertscore
- perplexity
---

# Model Card for Model ID

Fine-tuned using QLoRA for story generation task.


### Model Description

We utilize "Hierarchical Neural Story Generation" dataset and fine-tune the model to generate stories.

The input to the model is structred as follows:

'''

\#\#\# Instruction:  Below is a story idea. Write a short story based on this context.

\#\#\# Input: [story idea here]

\#\#\# Response:

'''


- **Developed by:** Abdelrahman ’Boda’ Sadallah, Anastasiia Demidova, Daria Kotova
- **Model type:** Causal LM
- **Language(s) (NLP):** English
- **Finetuned from model [optional]:** meta-llama/Llama-2-7b-hf

### Model Sources [optional]

<!-- Provide the basic links for the model. -->

- **Repository:** https://github.com/BodaSadalla98/llm-optimized-fintuning

## Uses

<!-- Address questions around how the model is intended to be used, including the foreseeable users of the model and those affected by the model. -->

The model is the result of our AI project. If you intend to use it, please, refer to the repo.


### Recommendations

<!-- This section is meant to convey recommendations with respect to the bias, risk, and technical limitations. -->

For improving stories generation, you can play parameters: temeperature, top_p/top_k, repetition_penalty, etc.

## How to Get Started with the Model

Use the code below to get started with the model.

[More Information Needed]

## Training Details

### Training Data

<!-- This should link to a Data Card, perhaps with a short stub of information on what the training data is all about as well as documentation related to data pre-processing or additional filtering. -->

Github for the dataset: https://github.com/kevalnagda/StoryGeneration

### Training Procedure 

<!-- This relates heavily to the Technical Specifications. Content here should link to that section when it is relevant to the training procedure. -->


#### Training Hyperparameters

- **Training regime:** [More Information Needed] <!--fp32, fp16 mixed precision, bf16 mixed precision, bf16 non-mixed precision, fp16 non-mixed precision, fp8 mixed precision -->

#### Speeds, Sizes, Times [optional]

<!-- This section provides information about throughput, start/end time, checkpoint size if relevant, etc. -->

[More Information Needed]

## Evaluation

<!-- This section describes the evaluation protocols and provides the results. -->

### Testing Data, Factors & Metrics

#### Testing Data

<!-- This should link to a Data Card if possible. -->

Test split of the same dataset.

#### Metrics

<!-- These are the evaluation metrics being used, ideally with a description of why. -->

We are using perplexity and BERTScore.

### Results

Perplexity: 8.0546

BERTScore: 80.11

#### Summary



## Model Examination [optional]

<!-- Relevant interpretability work for the model goes here -->

[More Information Needed]

## Environmental Impact

<!-- Total emissions (in grams of CO2eq) and additional considerations, such as electricity usage, go here. Edit the suggested text below accordingly -->

Carbon emissions can be estimated using the [Machine Learning Impact calculator](https://mlco2.github.io/impact#compute) presented in [Lacoste et al. (2019)](https://arxiv.org/abs/1910.09700).

- **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]


## Training procedure


The following `bitsandbytes` quantization config was used during training:
- quant_method: bitsandbytes
- load_in_8bit: False
- load_in_4bit: True
- llm_int8_threshold: 6.0
- llm_int8_skip_modules: None
- llm_int8_enable_fp32_cpu_offload: False
- llm_int8_has_fp16_weight: False
- bnb_4bit_quant_type: fp4
- bnb_4bit_use_double_quant: False
- bnb_4bit_compute_dtype: float32

### Framework versions


- PEFT 0.6.0.dev0