nm-testing/Llama-3.2-1B-Instruct-GPTQ-nonuniform

Creation

from datasets import load_dataset
from transformers import AutoTokenizer

from llmcompressor.transformers import SparseAutoModelForCausalLM, oneshot

# Select model and load it.
MODEL_ID = "meta-llama/Llama-3.2-1B-Instruct"

model = SparseAutoModelForCausalLM.from_pretrained(
    MODEL_ID,
    device_map="auto",
    torch_dtype="auto",
)
tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)

# Select calibration dataset.
DATASET_ID = "mgoin/ultrachat_2k"
DATASET_SPLIT = "train_sft"

# Select number of samples. 512 samples is a good place to start.
# Increasing the number of samples can improve accuracy.
NUM_CALIBRATION_SAMPLES = 512
MAX_SEQUENCE_LENGTH = 2048

# Load dataset and preprocess.
ds = load_dataset(DATASET_ID, split=DATASET_SPLIT)
ds = ds.shuffle(seed=42).select(range(NUM_CALIBRATION_SAMPLES))


def preprocess(example):
    return {
        "text": tokenizer.apply_chat_template(
            example["messages"],
            tokenize=False,
        )
    }


ds = ds.map(preprocess)


# Tokenize inputs.
def tokenize(sample):
    return tokenizer(
        sample["text"],
        padding=False,
        max_length=MAX_SEQUENCE_LENGTH,
        truncation=True,
        add_special_tokens=False,
    )


ds = ds.map(tokenize, remove_columns=ds.column_names)

# Configure the quantization algorithm to run.
recipe = """
quantization_stage:
  run_type: oneshot
  quantization_modifiers:
    GPTQModifier:
      ignore: ["lm_head"]
      config_groups:
        group_0:
          weights:
            num_bits: 4
            type: "int"
            symmetric: true
            strategy: "group"
            group_size: 128
          targets: [
                "re:.*mlp.[^.]*_proj",
          ]
        group_1:
          weights:
            num_bits: 8
            type: "int"
            symmetric: true
            strategy: "channel"
          targets: [
                "re:.*self_attn.[^.]*_proj",
          ]
"""

# Apply algorithms.
oneshot(
    model=model,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
)

# Save to disk compressed.
SAVE_DIR = MODEL_ID.split("/")[1] + "-GPTQ-nonuniform"
model.save_pretrained(SAVE_DIR)
tokenizer.save_pretrained(SAVE_DIR)

# Confirm generations of the quantized model look sane.
print("\n\n")
print("========== SAMPLE GENERATION ==============")
input_ids = tokenizer("Hello my name is", return_tensors="pt").input_ids.to("cuda")
output = model.generate(input_ids, max_new_tokens=100)
print(tokenizer.decode(output[0]))
print("==========================================\n\n")