---
library_name: transformers
license: apache-2.0
base_model: google/flan-t5-base
tags:
- generated_from_trainer
model-index:
- name: flan-t5-base-gen-12-small_dataset
  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. -->

[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/greatakela/gen_chatbot_models/runs/g7jitsum)
# flan-t5-base-gen-12-small_dataset

This model is a fine-tuned version of [google/flan-t5-base](https://huggingface.co/google/flan-t5-base) on the None dataset.
It achieves the following results on the evaluation set:
- Loss: 3.1621
- Rouge 1: 7.3814
- Rouge 2: 0.6192
- Rouge L: 6.8531
- Avg Len: 13.0278
- Bertscore Prec: 0.8612
- Bertscore Rec: 0.8542
- Bertscore F1: 0.8573

## 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
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 12

### Training results

| Training Loss | Epoch   | Step | Validation Loss | Rouge 1 | Rouge 2 | Rouge L | Avg Len | Bertscore Prec | Bertscore Rec | Bertscore F1 |
|:-------------:|:-------:|:----:|:---------------:|:-------:|:-------:|:-------:|:-------:|:--------------:|:-------------:|:------------:|
| 3.8134        | 0.6173  | 200  | 3.4410          | 6.2979  | 0.223   | 5.7832  | 13.5052 | 0.8507         | 0.8498        | 0.8498       |
| 3.5423        | 1.2346  | 400  | 3.3112          | 6.0189  | 0.3369  | 5.6265  | 14.6944 | 0.8611         | 0.8514        | 0.8558       |
| 3.3863        | 1.8519  | 600  | 3.2457          | 5.8478  | 0.312   | 5.5206  | 14.901  | 0.8649         | 0.8522        | 0.8581       |
| 3.2873        | 2.4691  | 800  | 3.2077          | 6.1468  | 0.4176  | 5.7813  | 14.4757 | 0.8643         | 0.8522        | 0.8578       |
| 3.2097        | 3.0864  | 1000 | 3.1873          | 6.8407  | 0.5555  | 6.391   | 13.6875 | 0.8591         | 0.8521        | 0.8553       |
| 3.1199        | 3.7037  | 1200 | 3.1723          | 6.6644  | 0.3774  | 6.2188  | 15.6545 | 0.8557         | 0.8511        | 0.8531       |
| 3.0885        | 4.3210  | 1400 | 3.1635          | 7.0627  | 0.5238  | 6.5367  | 14.4826 | 0.861          | 0.8527        | 0.8565       |
| 3.033         | 4.9383  | 1600 | 3.1565          | 7.0399  | 0.5467  | 6.4524  | 14.401  | 0.8596         | 0.8527        | 0.8558       |
| 2.9712        | 5.5556  | 1800 | 3.1555          | 7.1467  | 0.5327  | 6.4363  | 14.6406 | 0.8566         | 0.853         | 0.8545       |
| 2.9196        | 6.1728  | 2000 | 3.1563          | 7.1535  | 0.4741  | 6.6271  | 14.8073 | 0.8558         | 0.8531        | 0.8542       |
| 2.8896        | 6.7901  | 2200 | 3.1531          | 7.1215  | 0.5534  | 6.5025  | 14.408  | 0.8579         | 0.853         | 0.8551       |
| 2.8631        | 7.4074  | 2400 | 3.1547          | 7.4895  | 0.7019  | 6.8118  | 14.092  | 0.8581         | 0.8533        | 0.8554       |
| 2.8525        | 8.0247  | 2600 | 3.1532          | 7.1931  | 0.6333  | 6.6858  | 13.9201 | 0.8586         | 0.8528        | 0.8553       |
| 2.7951        | 8.6420  | 2800 | 3.1546          | 7.2016  | 0.7094  | 6.6671  | 13.4878 | 0.8599         | 0.8534        | 0.8563       |
| 2.7996        | 9.2593  | 3000 | 3.1568          | 7.225   | 0.6035  | 6.7029  | 13.724  | 0.8582         | 0.8532        | 0.8554       |
| 2.7721        | 9.8765  | 3200 | 3.1563          | 7.0646  | 0.6486  | 6.5622  | 13.125  | 0.8602         | 0.853         | 0.8562       |
| 2.759         | 10.4938 | 3400 | 3.1625          | 7.3836  | 0.7279  | 6.9035  | 12.6927 | 0.8613         | 0.8535        | 0.857        |
| 2.7459        | 11.1111 | 3600 | 3.1600          | 7.4314  | 0.6359  | 6.8986  | 13.1528 | 0.8605         | 0.8539        | 0.8569       |
| 2.7356        | 11.7284 | 3800 | 3.1621          | 7.3814  | 0.6192  | 6.8531  | 13.0278 | 0.8612         | 0.8542        | 0.8573       |


### Framework versions

- Transformers 4.49.0
- Pytorch 2.6.0+cu124
- Datasets 3.4.1
- Tokenizers 0.21.1