pip install flash-attn

aitraining llm --train \
  --model meta-llama/Llama-3.2-1B \
  --data-path ./data.jsonl \
  --project-name fast-model \
  --use-flash-attention-2

from autotrain.trainers.clm.params import LLMTrainingParams

params = LLMTrainingParams(
    model="meta-llama/Llama-3.2-1B",
    data_path="./data.jsonl",
    project_name="fast-model",

    use_flash_attention_2=True,
)

params = LLMTrainingParams(
    model="meta-llama/Llama-3.2-8B",
    data_path="./data.jsonl",
    project_name="fast-quantized",

    peft=True,
    quantization="int4",
    use_flash_attention_2=True,
)

aitraining llm --train \
  --model meta-llama/Llama-3.2-8B \
  --data-path ./data.jsonl \
  --project-name fast-quantized \
  --peft \
  --quantization int4 \
  --use-flash-attention-2

params = LLMTrainingParams(
    model="meta-llama/Llama-3.2-1B",
    data_path="./data.jsonl",
    project_name="packed-model",

    use_flash_attention_2=True,
    packing=True,
)

# Certifique-se de que o toolkit CUDA está instalado
nvcc --version

# Instalar com versão CUDA específica
pip install flash-attn --no-build-isolation