---
library_name: peft
tags:
- trl
- sft
- generated_from_trainer
datasets:
- dialogstudio
base_model: NousResearch/Llama-2-7b-hf
model-index:
- name: llama2-7b-tweet-summarization
  results: []
---

<!-- This model card has been generated automatically according to the information the Trainer had access to. You
should probably proofread and complete it, then remove this comment. -->

# llama2-7b-tweet-summarization

This model is a fine-tuned version of [NousResearch/Llama-2-7b-hf](https://huggingface.co/NousResearch/Llama-2-7b-hf) on the dialogstudio dataset.
It achieves the following results on the evaluation set:
- Loss: 2.8876
- Rouge Scores: {'rouge1': 71.60622394148149, 'rouge2': 59.01568798771897, 'rougeL': 48.52977346441143, 'rougeLsum': 71.51397796132038}
- Bleu Scores: [0.663980810175294, 0.6539436277190132, 0.6336458977384846, 0.6103966849577666]
- Gen Len: 463.0182

## Model description

More information needed

## Intended uses & limitations

More information needed

## Training and evaluation data

More information needed

## Training procedure

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 4
- eval_batch_size: 8
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- num_epochs: 7
- mixed_precision_training: Native AMP

### Training results

| Training Loss | Epoch | Step | Validation Loss | Rouge Scores                                                                                                            | Bleu Scores                                                                      | Gen Len  |
|:-------------:|:-----:|:----:|:---------------:|:-----------------------------------------------------------------------------------------------------------------------:|:--------------------------------------------------------------------------------:|:--------:|
| 1.9342        | 1.0   | 220  | 1.8650          | {'rouge1': 81.3543855201909, 'rouge2': 68.33628607515259, 'rougeL': 58.82060003710821, 'rougeLsum': 81.31170089312123}  | [0.7475269615217999, 0.7320503944473035, 0.7092142839754116, 0.6841937810377433] | 463.0182 |
| 1.6847        | 2.0   | 440  | 1.8651          | {'rouge1': 82.24291097248478, 'rouge2': 69.08611946264551, 'rougeL': 59.03857240450188, 'rougeLsum': 82.21842705467084} | [0.7506228265880205, 0.7370025856405312, 0.7147650770848061, 0.6900216562765312] | 463.0182 |
| 1.3489        | 3.0   | 660  | 1.9777          | {'rouge1': 74.8813002669941, 'rouge2': 62.506440459770204, 'rougeL': 52.7220886983953, 'rougeLsum': 74.84300846787131}  | [0.6692327330499573, 0.6584063380227538, 0.6386775472556582, 0.6163741473600254] | 463.0182 |
| 0.9474        | 4.0   | 880  | 2.1929          | {'rouge1': 78.71628333472167, 'rouge2': 65.58881455717321, 'rougeL': 55.06014924478776, 'rougeLsum': 78.69638389807935} | [0.7102475447746652, 0.6987491736376487, 0.6776051942379059, 0.6536770956211561] | 463.0182 |
| 0.6111        | 5.0   | 1100 | 2.4439          | {'rouge1': 78.27895756478365, 'rouge2': 64.90905260843559, 'rougeL': 54.06285449810264, 'rougeLsum': 78.22155552696411} | [0.7449392180339443, 0.7329638020466664, 0.7106292329793894, 0.6852866075200218] | 463.0182 |
| 0.3982        | 6.0   | 1320 | 2.7334          | {'rouge1': 70.6225818335813, 'rouge2': 58.36070884299628, 'rougeL': 48.1310974990119, 'rougeLsum': 70.48308395430783}   | [0.6525378114736703, 0.6423715222547518, 0.6226082916386746, 0.6000829162052789] | 463.0182 |
| 0.3039        | 7.0   | 1540 | 2.8876          | {'rouge1': 71.60622394148149, 'rouge2': 59.01568798771897, 'rougeL': 48.52977346441143, 'rougeLsum': 71.51397796132038} | [0.663980810175294, 0.6539436277190132, 0.6336458977384846, 0.6103966849577666]  | 463.0182 |


### Framework versions

- PEFT 0.8.2.dev0
- Transformers 4.38.0.dev0
- Pytorch 2.1.0+cu121
- Datasets 2.16.2.dev0
- Tokenizers 0.15.1