---
license: other
library_name: peft
tags:
- generated_from_trainer
base_model: google/gemma-7b
metrics:
- accuracy
model-index:
- name: lex_glue_ledgar
  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. -->

# lex_glue_ledgar

This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.5041
- Accuracy: 0.8662
- F1 Macro: 0.7935
- F1 Micro: 0.8662

## 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: 5e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- total_train_batch_size: 32
- total_eval_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 3.0

### Training results

| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Micro |
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:--------:|
| 1.3725        | 0.05  | 100  | 1.3878          | 0.6864   | 0.5157   | 0.6864   |
| 1.3256        | 0.11  | 200  | 1.0876          | 0.7615   | 0.6078   | 0.7615   |
| 0.9681        | 0.16  | 300  | 0.9516          | 0.7699   | 0.6452   | 0.7699   |
| 0.9094        | 0.21  | 400  | 0.9403          | 0.7893   | 0.6628   | 0.7893   |
| 0.7715        | 0.27  | 500  | 0.8593          | 0.7896   | 0.6687   | 0.7896   |
| 0.7244        | 0.32  | 600  | 0.7621          | 0.8061   | 0.6949   | 0.8061   |
| 0.7719        | 0.37  | 700  | 0.8355          | 0.7884   | 0.6864   | 0.7884   |
| 0.6305        | 0.43  | 800  | 0.8542          | 0.7897   | 0.6807   | 0.7897   |
| 0.8793        | 0.48  | 900  | 0.8043          | 0.7935   | 0.6822   | 0.7935   |
| 0.7411        | 0.53  | 1000 | 0.7256          | 0.8072   | 0.6940   | 0.8072   |
| 0.6403        | 0.59  | 1100 | 0.7033          | 0.819    | 0.7217   | 0.819    |
| 0.6971        | 0.64  | 1200 | 0.7009          | 0.8159   | 0.7335   | 0.8159   |
| 0.7053        | 0.69  | 1300 | 0.6921          | 0.8291   | 0.7205   | 0.8291   |
| 0.6413        | 0.75  | 1400 | 0.6515          | 0.8301   | 0.7292   | 0.8301   |
| 0.6656        | 0.8   | 1500 | 0.6685          | 0.8241   | 0.7161   | 0.8241   |
| 0.6114        | 0.85  | 1600 | 0.6453          | 0.8246   | 0.7269   | 0.8246   |
| 0.5616        | 0.91  | 1700 | 0.6632          | 0.8275   | 0.7290   | 0.8275   |
| 0.6985        | 0.96  | 1800 | 0.6022          | 0.8329   | 0.7395   | 0.8329   |
| 0.387         | 1.01  | 1900 | 0.5910          | 0.8475   | 0.7690   | 0.8475   |
| 0.2391        | 1.07  | 2000 | 0.6235          | 0.8475   | 0.7564   | 0.8475   |
| 0.4414        | 1.12  | 2100 | 0.6027          | 0.8421   | 0.7651   | 0.8421   |
| 0.3869        | 1.17  | 2200 | 0.6028          | 0.8437   | 0.7592   | 0.8437   |
| 0.2387        | 1.23  | 2300 | 0.6646          | 0.845    | 0.7635   | 0.845    |
| 0.3556        | 1.28  | 2400 | 0.6032          | 0.8487   | 0.7724   | 0.8487   |
| 0.4439        | 1.33  | 2500 | 0.5773          | 0.8589   | 0.7790   | 0.8589   |
| 0.4171        | 1.39  | 2600 | 0.5602          | 0.8551   | 0.7760   | 0.8551   |
| 0.3984        | 1.44  | 2700 | 0.5800          | 0.8514   | 0.7708   | 0.8514   |
| 0.2491        | 1.49  | 2800 | 0.5934          | 0.8463   | 0.7774   | 0.8463   |
| 0.2975        | 1.55  | 2900 | 0.5838          | 0.8548   | 0.7776   | 0.8548   |
| 0.4375        | 1.6   | 3000 | 0.5584          | 0.8497   | 0.7758   | 0.8497   |
| 0.3108        | 1.65  | 3100 | 0.5625          | 0.8624   | 0.7864   | 0.8624   |
| 0.3546        | 1.71  | 3200 | 0.5264          | 0.8586   | 0.7814   | 0.8586   |
| 0.4125        | 1.76  | 3300 | 0.5484          | 0.8509   | 0.7788   | 0.8509   |
| 0.2206        | 1.81  | 3400 | 0.5634          | 0.8563   | 0.7800   | 0.8563   |
| 0.3348        | 1.87  | 3500 | 0.5154          | 0.8644   | 0.7890   | 0.8644   |
| 0.3451        | 1.92  | 3600 | 0.5221          | 0.8667   | 0.7858   | 0.8667   |
| 0.3077        | 1.97  | 3700 | 0.5041          | 0.8662   | 0.7935   | 0.8662   |
| 0.1352        | 2.03  | 3800 | 0.5687          | 0.8668   | 0.7919   | 0.8668   |
| 0.1012        | 2.08  | 3900 | 0.5754          | 0.8651   | 0.7888   | 0.8651   |
| 0.1006        | 2.13  | 4000 | 0.5929          | 0.872    | 0.7959   | 0.872    |
| 0.0536        | 2.19  | 4100 | 0.5760          | 0.8739   | 0.7992   | 0.8739   |
| 0.0401        | 2.24  | 4200 | 0.6251          | 0.87     | 0.7935   | 0.87     |
| 0.0756        | 2.29  | 4300 | 0.5895          | 0.8709   | 0.8027   | 0.8709   |
| 0.0501        | 2.35  | 4400 | 0.5434          | 0.8707   | 0.7962   | 0.8707   |
| 0.0611        | 2.4   | 4500 | 0.5949          | 0.8759   | 0.8042   | 0.8759   |
| 0.081         | 2.45  | 4600 | 0.6089          | 0.8787   | 0.8122   | 0.8787   |
| 0.1033        | 2.51  | 4700 | 0.5790          | 0.8752   | 0.8107   | 0.8752   |
| 0.1131        | 2.56  | 4800 | 0.5828          | 0.8747   | 0.8036   | 0.8747   |
| 0.094         | 2.61  | 4900 | 0.5612          | 0.878    | 0.8107   | 0.878    |
| 0.0853        | 2.67  | 5000 | 0.5772          | 0.8784   | 0.8123   | 0.8784   |
| 0.0917        | 2.72  | 5100 | 0.5595          | 0.8805   | 0.8123   | 0.8805   |
| 0.0542        | 2.77  | 5200 | 0.5782          | 0.8814   | 0.8147   | 0.8814   |
| 0.0754        | 2.83  | 5300 | 0.5936          | 0.8821   | 0.8171   | 0.8821   |
| 0.1001        | 2.88  | 5400 | 0.5626          | 0.8827   | 0.8157   | 0.8827   |
| 0.0311        | 2.93  | 5500 | 0.5690          | 0.8818   | 0.8152   | 0.8818   |
| 0.03          | 2.99  | 5600 | 0.5688          | 0.8831   | 0.8171   | 0.8831   |


### Framework versions

- PEFT 0.9.0
- Transformers 4.39.0.dev0
- Pytorch 2.2.1+cu121
- Datasets 2.18.0
- Tokenizers 0.15.2