---
library_name: transformers
license: apache-2.0
base_model: Qwen/Qwen3-0.6B
tags:
- generated_from_trainer
model-index:
- name: Qwen3-thirukkural-tamil-v2
  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. -->

# Qwen3-thirukkural-tamil-v2

This model is a fine-tuned version of [Qwen/Qwen3-0.6B](https://huggingface.co/Qwen/Qwen3-0.6B) on an [Thirukkural dataset](https://huggingface.co/datasets/aitamilnadu/thirukkural_instruct).
It achieves the following results on the evaluation set:
- Loss: 0.2986

### Training hyperparameters

The following hyperparameters were used during training:
- learning_rate: 2e-05
- train_batch_size: 2
- eval_batch_size: 1
- seed: 42
- gradient_accumulation_steps: 8
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- num_epochs: 6

### Training results

| Training Loss | Epoch | Step | Validation Loss |
|:-------------:|:-----:|:----:|:---------------:|
| 0.2331        | 1.0   | 200  | 0.3307          |
| 0.1840        | 2.0   | 400  | 0.3093          |
| 0.1735        | 3.0   | 600  | 0.3013          |
| 0.1550        | 4.0   | 800  | 0.2983          |
| 0.1789        | 5.0   | 1000 | 0.2980          |
| 0.1689        | 6.0   | 1200 | 0.2986          |

### Usage


``` Python

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "tidelganesh/Qwen3-thirukkural-tamil"  # your repo
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, dtype="auto")

messages = [{"role": "user", "content": "பொறையுடைமை அதிகாரத்தில் வரும் 158ஆம் குறளைத் தருக."}]
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
inputs = tokenizer(prompt, return_tensors="pt").to(model.device)

output = model.generate(
    **inputs, max_new_tokens=150,
    eos_token_id=tokenizer.eos_token_id,
    pad_token_id=tokenizer.eos_token_id,
)
print(tokenizer.decode(output[0][inputs["input_ids"].shape[1]:], skip_special_tokens=True))
```
### Framework versions

- Transformers 5.14.1
- Pytorch 2.8.0+cu128
- Datasets 5.0.0
- Tokenizers 0.22.2