metadata
library_name: peft
base_model: sarvamai/sarvam-1
tags:
- generated_from_trainer
model-index:
- name: Hindi_Normalisation
results: []
Hindi_Normalisation
This model is a fine-tuned version of sarvamai/sarvam-1 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.9944
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: 1
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- 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: cosine
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.7981 | 0.25 | 50 | 3.6827 |
| 3.5768 | 0.5 | 100 | 3.5449 |
| 3.2662 | 0.75 | 150 | 3.2327 |
| 2.9061 | 1.0 | 200 | 2.9117 |
| 2.6027 | 1.25 | 250 | 2.6400 |
| 2.4047 | 1.5 | 300 | 2.3961 |
| 2.1897 | 1.75 | 350 | 2.1825 |
| 2.0345 | 2.0 | 400 | 2.1042 |
| 2.0123 | 2.25 | 450 | 2.0725 |
| 2.0704 | 2.5 | 500 | 2.0525 |
| 1.9966 | 2.75 | 550 | 2.0358 |
| 2.0171 | 3.0 | 600 | 2.0233 |
| 1.9052 | 3.25 | 650 | 2.0137 |
| 2.0293 | 3.5 | 700 | 2.0067 |
| 1.961 | 3.75 | 750 | 2.0017 |
| 1.9371 | 4.0 | 800 | 1.9979 |
| 2.006 | 4.25 | 850 | 1.9959 |
| 2.0385 | 4.5 | 900 | 1.9948 |
| 1.9319 | 4.75 | 950 | 1.9945 |
| 1.9252 | 5.0 | 1000 | 1.9944 |
Framework versions
- PEFT 0.14.0
- Transformers 4.48.2
- Pytorch 2.5.1+cu124
- Datasets 3.2.0
- Tokenizers 0.21.0