MT_En_Hindi

import torch
from transformers import AutoTokenizer, AutoModelForCausalLM

# Load tokenizer and model
model_name = "baban/MT_En_Hindi" 
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.bfloat16, device_map="auto")

source_text = "The weather is nice today."
prompt = f"Translate the following English sentence to Hindi:\n{source_text}"

messages = [
    {"role": "user", "content": prompt}
]

# Tokenize the formatted input
input_ids = tokenizer.apply_chat_template(messages, return_tensors="pt").to(model.device)


with torch.no_grad():
    output_ids = model.generate(
      input_ids=input_ids,
      max_new_tokens=100,
      do_sample=False
  )


# Decode and print only the new tokens (the response)
response = tokenizer.decode(output_ids[0][input_ids.shape[-1]:], skip_special_tokens=True)
print("\n=== Translation ===")
print(response)

<!-- 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. -->

MT_En_Hindi

This model is a fine-tuned version of meta-llama/Llama-3.2-1B on the MT_En_Hindi dataset. It achieves the following results on the evaluation set:

• Loss: 0.5924
• Num Input Tokens Seen: 6566229120

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: 8
• gradient_accumulation_steps: 8
• total_train_batch_size: 1024
• total_eval_batch_size: 128
• 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: inverse_sqrt
• num_epochs: 3.0

Training results

Framework versions

• Transformers 4.52.4
• Pytorch 2.7.0+cu128
• Datasets 3.6.0
• Tokenizers 0.21.1