tags:

• fp8
• vllm language:
• en
• de
• fr
• it
• pt
• hi
• es
• th pipeline_tag: text-generation license: llama3.1 base_model: meta-llama/Meta-Llama-3.1-70B

Meta-Llama-3.1-70B-FP8

Model Overview

• Model Architecture: Meta-Llama-3.1
  • Input: Text
  • Output: Text
• Model Optimizations:
  • Weight quantization: FP8
  • Activation quantization: FP8
• Intended Use Cases: Intended for commercial and research use in multiple languages. Similarly to Meta-Llama-3.1-8B, this model serves as a base version.
• Out-of-scope: Use in any manner that violates applicable laws or regulations (including trade compliance laws). Use in languages other than English.
• Release Date: 7/23/2024
• Version: 1.0
• License(s): llama3.1
• Model Developers: Neural Magic

Quantized version of Meta-Llama-3.1-70B. It achieves an average score of 79.70 on the OpenLLM benchmark (version 1), whereas the unquantized model achieves 79.84.

Model Optimizations

This model was obtained by quantizing the weights and activations of Meta-Llama-3.1-70B to FP8 data type, ready for inference with vLLM built from source. This optimization reduces the number of bits per parameter from 16 to 8, reducing the disk size and GPU memory requirements by approximately 50%.

Only the weights and activations of the linear operators within transformers blocks are quantized. Symmetric per-tensor quantization is applied, in which a single linear scaling maps the FP8 representations of the quantized weights and activations. LLM Compressor is used for quantization with 512 sequences of UltraChat.

Creation

This model was created by applying LLM Compressor with calibration samples from UltraChat, as presented in the code snipet below.

import torch
from datasets import load_dataset
from transformers import AutoTokenizer

from llmcompressor.transformers import SparseAutoModelForCausalLM, oneshot
from llmcompressor.transformers.compression.helpers import (
    calculate_offload_device_map,
    custom_offload_device_map,
)

recipe = """
quant_stage:
    quant_modifiers:
        QuantizationModifier:
            ignore: ["lm_head"]
            config_groups:
                group_0:
                    weights:
                        num_bits: 8
                        type: float
                        strategy: tensor
                        dynamic: false
                        symmetric: true
                    input_activations:
                        num_bits: 8
                        type: float
                        strategy: tensor
                        dynamic: false
                        symmetric: true
                    targets: ["Linear"]
"""

model_stub = "meta-llama/Meta-Llama-3.1-70B"
model_name = model_stub.split("/")[-1]

device_map = calculate_offload_device_map(
    model_stub, reserve_for_hessians=False, num_gpus=2, torch_dtype=torch.float16
)

model = SparseAutoModelForCausalLM.from_pretrained(
    model_stub, torch_dtype=torch.float16, device_map=device_map
)
tokenizer = AutoTokenizer.from_pretrained(model_stub)

output_dir = f"./{model_name}-FP8"

DATASET_ID = "HuggingFaceH4/ultrachat_200k"
DATASET_SPLIT = "train_sft"
NUM_CALIBRATION_SAMPLES = 512
MAX_SEQUENCE_LENGTH = 4096

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)

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)

oneshot(
    model=model,
    output_dir=output_dir,
    dataset=ds,
    recipe=recipe,
    max_seq_length=MAX_SEQUENCE_LENGTH,
    num_calibration_samples=NUM_CALIBRATION_SAMPLES,
    save_compressed=True,
)

Evaluation

The model was evaluated on MMLU, ARC-Challenge, GSM-8K, Hellaswag, Winogrande and TruthfulQA. Evaluation was conducted using the Neural Magic fork of lm-evaluation-harness (branch llama_3.1_instruct) and the vLLM engine. This version of the lm-evaluation-harness includes versions of ARC-Challenge that matches the prompting style of Meta-Llama-3.1-evals.

Accuracy

Open LLM Leaderboard evaluation scores

Reproduction

The results were obtained using the following commands:

MMLU

lm_eval \
  --model vllm \
  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-FP8",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=2 \
  --tasks mmlu \
  --num_fewshot 5 \
  --batch_size auto

ARC-Challenge

lm_eval \
  --model vllm \
  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-FP8",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=2 \
  --tasks arc_challenge_llama_3.1_instruct \
  --num_fewshot 25 \
  --batch_size auto

GSM-8K

lm_eval \
  --model vllm \
  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-FP8",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=2 \
  --tasks gsm8k \
  --num_fewshot 5 \
  --batch_size auto

Hellaswag

lm_eval \
  --model vllm \
  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-FP8",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=2 \
  --tasks hellaswag \
  --num_fewshot 10 \
  --batch_size auto

Winogrande

lm_eval \
  --model vllm \
  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-FP8",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=2 \
  --tasks winogrande \
  --num_fewshot 5 \
  --batch_size auto

TruthfulQA

lm_eval \
  --model vllm \
  --model_args pretrained="neuralmagic/Meta-Llama-3.1-70B-FP8",dtype=auto,add_bos_token=True,max_model_len=4096,tensor_parallel_size=2 \
  --tasks truthfulqa_mc \
  --num_fewshot 0 \
  --batch_size auto