mariamffatima/Pakistan-Legal-ChatBot

1---2library_name: peft3license: apache-2.04base_model: google/flan-t5-base5tags:6- base_model:adapter:google/flan-t5-base7- lora8- transformers9metrics:10- rouge11model-index:12- name: Pakistan-Legal-ChatBot13  results: []14---15 16<!-- This model card has been generated automatically according to the information the Trainer had access to. You17should probably proofread and complete it, then remove this comment. -->18 19# Pakistan-Legal-ChatBot20 21This model is a fine-tuned version of [google/flan-t5-base](https://huggingface.co/google/flan-t5-base) on an unknown dataset.22It achieves the following results on the evaluation set:23- Loss: nan24- Rouge1: 0.236525- Rouge2: 0.090726- Rougel: 0.190527- Rougelsum: 0.190628 29## Model description30 31More information needed32 33## Intended uses & limitations34 35More information needed36 37## Training and evaluation data38 39More information needed40 41## Training procedure42 43### Training hyperparameters44 45The following hyperparameters were used during training:46- learning_rate: 0.000547- train_batch_size: 848- eval_batch_size: 849- seed: 4250- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments51- lr_scheduler_type: linear52- num_epochs: 553- mixed_precision_training: Native AMP54 55### Training results56 57| Training Loss | Epoch | Step | Validation Loss | Rouge1 | Rouge2 | Rougel | Rougelsum |58|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:---------:|59| No log        | 1.0   | 245  | nan             | 0.2365 | 0.0907 | 0.1905 | 0.1906    |60| No log        | 2.0   | 490  | nan             | 0.2365 | 0.0907 | 0.1905 | 0.1906    |61| 0.0           | 3.0   | 735  | nan             | 0.2365 | 0.0907 | 0.1905 | 0.1906    |62| 0.0           | 4.0   | 980  | nan             | 0.2365 | 0.0907 | 0.1905 | 0.1906    |63| 0.0           | 5.0   | 1225 | nan             | 0.2365 | 0.0907 | 0.1905 | 0.1906    |64 65 66### Framework versions67 68- PEFT 0.18.069- Transformers 4.57.370- Pytorch 2.9.0+cu12671- Datasets 4.0.072- Tokenizers 0.22.2