smollm2-135m-raw-trained

This model is a custom trained version of SmolLM2-135M from scratch.

Model Details

• Base Model: SmolLM2-135M
• Parameters: ~135M
• Training: Custom trained using PyTorch Lightning

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("vikashkr117/smollm2-135m-raw-trained")
model = AutoModelForCausalLM.from_pretrained("vikashkr117/smollm2-135m-raw-trained")

# Generate text
input_text = "Never forget who you are,"
inputs = tokenizer(input_text, return_tensors="pt")
outputs = model.generate(**inputs, max_new_tokens=100, do_sample=True, temperature=0.7)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))