broadfield-dev/gemma-3-270m-context-qa

This model is a fine-tuned version of broadfield-dev/gemma-3-270m-tuned-0105-1022-tuned-0105-1045-tuned-0105-1111-tuned-0105-1133 on the broadfield-dev/deepmind_narrativeqa-Broadfield dataset.

Training Details

• Task: CAUSAL_LM
• Epochs: 1
• Learning Rate: 2e-05
• Gradient Accumulation Steps: 4

Entity Labels

['LABEL_0', 'LABEL_1']

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM
import torch
model_id = "broadfield-dev/gemma-3-270m-context-qa"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(model_id, torch_dtype=torch.float16)
messages = [
    {"role": "system", "content": "Using the context, answer the users question."},
    {"role": "user", "content": "Context: {context_content}\n\nQuestion: Your input here..."}

]
inputs = tokenizer.apply_chat_template(messages, return_tensors="pt", add_generation_prompt=True).to(model.device)
outputs = model.generate(inputs, max_new_tokens=256)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))