chunked/nltk_chunking/_quantization/chunk_89.txt

To find the best threshold for your model, we recommend experimenting with the llm_int8_threshold parameter in [BitsAndBytesConfig]:

from transformers import AutoModelForCausalLM, BitsAndBytesConfig
model_id = "bigscience/bloom-1b7"
quantization_config = BitsAndBytesConfig(
    llm_int8_threshold=10,
)
model_8bit = AutoModelForCausalLM.from_pretrained(
    model_id,
    device_map=device_map,
    quantization_config=quantization_config,
)

Skip module conversion
For some models, like Jukebox, you don't need to quantize every module to 8-bit which can actually cause instability.