metadata
extra_gated_heading: You need to share contact information with Meta to access this model
extra_gated_prompt: >-
## FAIR Noncommercial Research License
Last Updated: [October 16th 2024]
“Acceptable Use Policy” means the FAIR Acceptable Use Policy, applicable to
Research Materials, that is incorporated into this Agreement.
“Agreement” means the terms and conditions for use, reproduction, distribution
and modification of the Research Materials set forth herein.
“Documentation” means the specifications, manuals and documentation
accompanying Research Materials distributed by Meta.
“Licensee” or “you” means you, or your employer or any other person or entity
(if you are entering into this Agreement on such person or entity’s behalf),
of the age required under applicable laws, rules or regulations to provide
legal consent and that has legal authority to bind your employer or such other
person or entity if you are entering in this Agreement on their behalf.
“Meta” or “we” means Meta Platforms Ireland Limited (if you are located in or,
if you are an entity, your principal place of business is in the EEA or
Switzerland) and Meta Platforms, Inc. (if you are located outside of the EEA
or Switzerland).
“Noncommercial Research Uses” means noncommercial research use cases related
to research, development, education, processing, or analysis and in each case,
is not primarily intended for commercial advantage or monetary compensation to
you or others.
“Research Materials” means, collectively, Documentation and the models,
software and algorithms, including machine-learning model code, trained model
weights, inference-enabling code, training-enabling code, fine-tuning enabling
code, demonstration materials and other elements of the foregoing distributed
by Meta and made available under this Agreement.
By clicking “I Accept” below or by using or distributing any portion or
element of the Research Materials, you agree to be bound by this Agreement.
1. License Rights and Redistribution.
a. Grant of Rights. You are granted a non-exclusive, worldwide, non-transferable and royalty-free limited license under Meta’s intellectual property or other rights owned by Meta embodied in the Research Materials to use, reproduce, distribute, copy, create derivative works of, and make modifications to the Research Materials.
b. Redistribution and Use.
i. You will not use the Research Materials or any outputs or results of the Research Materials in connection with any commercial uses or for any uses other than Noncommercial Research Uses;
ii. Distribution of Research Materials, and any derivative works thereof, are subject to the terms of this Agreement. If you distribute or make the Research Materials, or any derivative works thereof, available to a third party, you may only do so under the terms of this Agreement. You shall also provide a copy of this Agreement to such third party.
iii. If you submit for publication the results of research you perform on, using, or otherwise in connection with Research Materials, you must acknowledge the use of Research Materials in your publication.
iv. Your use of the Research Materials must comply with applicable laws and regulations (including Trade Control Laws) and adhere to the FAIR Acceptable Use Policy, which is hereby incorporated by reference into this Agreement.
2. User Support. Your Noncommercial Research Use of the Research Materials is
done at your own discretion; Meta does not process any information nor provide
any service in relation to such use. Meta is under no obligation to provide
any support services for the Research Materials. Any support provided is “as
is”, “with all faults”, and without warranty of any kind.
3. Disclaimer of Warranty. UNLESS REQUIRED BY APPLICABLE LAW, THE RESEARCH
MATERIALS AND ANY OUTPUT AND RESULTS THEREFROM ARE PROVIDED ON AN “AS IS”
BASIS, WITHOUT WARRANTIES OF ANY KIND, AND META DISCLAIMS ALL WARRANTIES OF
ANY KIND, BOTH EXPRESS AND IMPLIED, INCLUDING, WITHOUT LIMITATION, ANY
WARRANTIES OF TITLE, NON-INFRINGEMENT, MERCHANTABILITY, OR FITNESS FOR A
PARTICULAR PURPOSE. YOU ARE SOLELY RESPONSIBLE FOR DETERMINING THE
APPROPRIATENESS OF USING OR REDISTRIBUTING THE RESEARCH MATERIALS AND ASSUME
ANY RISKS ASSOCIATED WITH YOUR USE OF THE RESEARCH MATERIALS AND ANY OUTPUT
AND RESULTS.
4. Limitation of Liability. IN NO EVENT WILL META OR ITS AFFILIATES BE LIABLE
UNDER ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, TORT, NEGLIGENCE, PRODUCTS
LIABILITY, OR OTHERWISE, ARISING OUT OF THIS AGREEMENT, FOR ANY LOST PROFITS
OR ANY DIRECT OR INDIRECT, SPECIAL, CONSEQUENTIAL, INCIDENTAL, EXEMPLARY OR
PUNITIVE DAMAGES, EVEN IF META OR ITS AFFILIATES HAVE BEEN ADVISED OF THE
POSSIBILITY OF ANY OF THE FOREGOING.
5. Intellectual Property.
a. Subject to Meta’s ownership of Research Materials and derivatives made by or for Meta, with respect to any derivative works and modifications of the Research Materials that are made by you, as between you and Meta, you are and will be the owner of such derivative works and modifications.
b. If you institute litigation or other proceedings against Meta or any entity (including a cross-claim or counterclaim in a lawsuit) alleging that the Research Materials, outputs or results, or any portion of any of the foregoing, constitutes infringement of intellectual property or other rights owned or licensable by you, then any licenses granted to you under this Agreement shall terminate as of the date such litigation or claim is filed or instituted. You will indemnify and hold harmless Meta from and against any claim by any third party arising out of or related to your use or distribution of the Research Materials.
6. Term and Termination. The term of this Agreement will commence upon your
acceptance of this Agreement or access to the Research Materials and will
continue in full force and effect until terminated in accordance with the
terms and conditions herein. Meta may terminate this Agreement if you are in
breach of any term or condition of this Agreement. Upon termination of this
Agreement, you shall delete and cease use of the Research Materials. Sections
5, 6 and 9 shall survive the termination of this Agreement.
7. Governing Law and Jurisdiction. This Agreement will be governed and
construed under the laws of the State of California without regard to choice
of law principles, and the UN Convention on Contracts for the International
Sale of Goods does not apply to this Agreement. The courts of California shall
have exclusive jurisdiction of any dispute arising out of this Agreement.
8. Modifications and Amendments. Meta may modify this Agreement from time to
time by posting a revised version at
https://huggingface.co/facebook/layerskip-llama2-7B/blob/main/LICENSE;
provided that they are similar in spirit to the current version of the
Agreement, but may differ in detail to address new problems or concerns. All
such changes will be effective immediately. Your continued use of the Research
Materials after any modification to this Agreement constitutes your agreement
to such modification. Except as provided in this Agreement, no modification or
addition to any provision of this Agreement will be binding unless it is in
writing and signed by an authorized representative of both you and Meta.
FAIR Acceptable Use Policy
The Fundamental AI Research (FAIR) team at Meta seeks to further understanding
of new and existing research domains with the mission of advancing the
state-of-the-art in artificial intelligence through open research for the
benefit of all.
As part of this mission, Meta makes certain research materials available for
noncommercial research use. Meta is committed to promoting the safe and
responsible use of such research materials.
Prohibited Uses
You agree you will not use, or allow others to use, Research Materials to:
1.Violate the law or others’ rights, including to:
a. Engage in, promote, generate, contribute to, encourage, plan, incite, or further illegal or unlawful activity or content, such as:
i. Violence or terrorism
ii. Exploitation or harm to children, including the solicitation, creation, acquisition, or dissemination of child exploitative content or failure to report Child Sexual Abuse Material
iii. Human trafficking, exploitation, and sexual violence
iv. The illegal distribution of information or materials to minors, including obscene materials, or failure to employ legally required age-gating in connection with such information or materials.
v. Sexual solicitation
vi. Any other criminal activity
b. Engage in, promote, incite, or facilitate the harassment, abuse, threatening, or bullying of individuals or groups of individuals
c. Engage in, promote, incite, or facilitate discrimination or other unlawful or harmful conduct in the provision of employment, employment benefits, credit, housing, other economic benefits, or other essential goods and services
d. Engage in the unauthorized or unlicensed practice of any profession including, but not limited to, financial, legal, medical/health, or related professional practices
e. Collect, process, disclose, generate, or infer health, demographic, or other sensitive personal or private information about individuals without rights and consents required by applicable laws
f. Engage in or facilitate any action or generate any content that infringes, misappropriates, or otherwise violates any third-party rights, including the outputs or results of any technology using FAIR research materials
g. Create, generate, or facilitate the creation of malicious code, malware, computer viruses or do anything else that could disable, overburden, interfere with or impair the proper working, integrity, operation or appearance of a website or computer system
2. Engage in, promote, incite, facilitate, or assist in the planning or
development of activities that present a risk of death or bodily harm to
individuals, including use of research artifacts related to the following:
a. Military, warfare, nuclear industries or applications, espionage, use for materials or activities that are subject to the International Traffic Arms Regulations (ITAR) maintained by the United States Department of State
b. Guns and illegal weapons (including weapon development)
c. Illegal drugs and regulated/controlled substances
d. Operation of critical infrastructure, transportation technologies, or heavy machinery
e. Self-harm or harm to others, including suicide, cutting, and eating disorders
f. Any content intended to incite or promote violence, abuse, or any infliction of bodily harm to an individual
3. Intentionally deceive or mislead others, including use of FAIR Research
Materials related to the following:
a. Generating, promoting, or furthering fraud or the creation or promotion of disinformation
b. Generating, promoting, or furthering defamatory content, including the creation of defamatory statements, images, or other content
c. Generating, promoting, or further distributing spam
d. Impersonating another individual without consent, authorization, or legal right
e. Representing that outputs of FAIR research materials or outputs from technology using FAIR research materials o are human-generated
f. Generating or facilitating false online engagement, including fake reviews and other means of fake online engagement
4. Fail to appropriately disclose to end users any known dangers of your
Research Materials.
Please report any violation of this Policy or other problems that could lead
to a violation of this Policy by submitting a report
[here](https://docs.google.com/forms/d/e/1FAIpQLSeb11cryAopJ7LNrC4nxEUXrHY26hfkXQMf_uH-oFgA3WlYZQ/viewform).
extra_gated_fields:
First Name: text
Last Name: text
Date of birth: date_picker
Country: country
Affiliation: text
geo: ip_location
By clicking Submit below I accept the terms of the license and acknowledge that the information I provide will be collected stored processed and shared in accordance with the Meta Privacy Policy: checkbox
extra_gated_description: >-
The information you provide will be collected, stored, processed and shared in
accordance with the [Meta Privacy
Policy](https://www.facebook.com/privacy/policy/).
extra_gated_button_content: Submit
language:
- en
library_name: transformers
pipeline_tag: text-generation
tags:
- facebook
- meta
- pytorch
- llama
- llama-2
model-index:
- name: LayerSkip Llama2 7B
results:
- task:
type: question-answering
dataset:
type: google/boolq
name: BoolQ
metrics:
- name: acc
type: acc
value: 0.776
verified: false
- task:
type: question-answering
dataset:
type: ybisk/piqa
name: PIQA
metrics:
- name: acc
type: acc
value: 0.775
verified: false
- task:
type: question-answering
dataset:
type: allenai/social_i_qa
name: SIQA
metrics:
- name: acc
type: acc
value: 0.454
verified: false
- task:
type: text-generation
dataset:
type: Rowan/hellaswag
name: HellaSwag
metrics:
- name: acc
type: acc
value: 0.567
verified: false
- task:
type: question-answering
dataset:
type: allenai/winogrande
name: WinoGrande
metrics:
- name: acc
type: acc
value: 0.701
verified: false
- task:
type: question-answering
dataset:
type: allenai/ai2_arc
name: ARC (Easy)
metrics:
- name: acc
type: acc
value: 0.765
verified: false
- task:
type: question-answering
dataset:
type: allenai/ai2_arc
name: ARC (Challenge)
metrics:
- name: acc
type: acc
value: 0.437
verified: false
- task:
type: question-answering
dataset:
type: allenai/openbookqa
name: OpenBookQA
metrics:
- name: acc
type: acc
value: 0.328
verified: false
- task:
type: question-answering
dataset:
type: ehovy/race
name: RACE
metrics:
- name: acc
type: acc
value: 0.389
verified: false
- task:
type: question-answering
dataset:
type: cais/mmlu
name: MMLU
metrics:
- name: acc
type: acc
value: 0.376
verified: false
- task:
type: text-generation
dataset:
type: google-research-datasets/nq_open
name: Natural Questions
metrics:
- name: exact_match
type: exact_match
value: 0.156
verified: false
- task:
type: question-answering
dataset:
type: sentence-transformers/trivia-qa
name: TriviaQA
metrics:
- name: acc
type: acc
value: 0.529
verified: false
- task:
type: text-generation
dataset:
type: openai/gsm8k
name: GSM8K
metrics:
- name: exact_match
type: exact_match
value: 0.121
verified: false
- task:
type: question-answering
dataset:
type: allenai/math_qa
name: MathQA
metrics:
- name: acc
type: acc
value: 0.276
verified: false
- task:
type: question-answering
dataset:
type: rajpurkar/squad_v2
name: SQuAD2.0
metrics:
- name: exact
type: exact
value: 0.164
verified: false
- task:
type: text-classification
dataset:
type: toxigen/toxigen-data
name: ToxiGen
metrics:
- name: acc
type: acc
value: 0.428
verified: false
- task:
type: text-generation
dataset:
type: openai_humaneval
name: HumanEval
metrics:
- name: pass@1
type: pass@1
value: 0.134
verified: false
- task:
type: text-generation
dataset:
type: mbpp
name: MBPP
metrics:
- name: pass@1
type: pass@1
value: 0.19
verified: false
license: other
license_name: fair
license_link: LICENSE
base_model: meta-llama/Llama-2-7b-hf
LayerSkip Llama2 7B
Llama2 7B model continually pretrained with LayerSkip as presented in Layer Skip: Enabling Early Exit Inference and Self-Speculative Decoding and is capable of performing self-speculative decoding: decode with earlier layers and verify with remaining layers.
How to Use
We are providing 3 ways to run the model
HuggingFace
HuggingFace does not yet have self-speculative decoding support. However, we can re-use it's speculative decoding feature by creating a draft model using a subset of the layers of the main model:
>>> from transformers import AutoModelForCausalLM, AutoTokenizer
>>> import torch
>>> from copy import deepcopy
>>> checkpoint = "facebook/layerskip-llama2-7B"
>>> early_exit = 4
>>> device = "cuda" if torch.cuda.is_available() else "cpu"
>>> prompt = "typing import List\ndef bucket_sort(A: List):"
>>> model = AutoModelForCausalLM.from_pretrained(checkpoint)
>>> tokenizer = AutoTokenizer.from_pretrained(checkpoint)
>>> generation_config = model.generation_config
>>> weights_memo = {id(w): w for w in model.parameters()}
>>> assistant_model = deepcopy(model, memo=weights_memo) # Clone main model with shared weights
>>> assistant_model.model.layers = assistant_model.model.layers[:early_exit] # Apply early exit
>>> del assistant_model.model.layers[early_exit:]
>>> model.to(device)
>>> assistant_model.to(device)
>>> inputs = tokenizer(prompt, return_tensors="pt").to(device)
>>> outputs = model.generate(**inputs, assistant_model=assistant_model, generation_config=generation_config)
>>> print(tokenizer.batch_decode(outputs, skip_special_tokens=True)[0])
Please note that this is not an optimal implementation as it requires more memory to save KV cache and activations of duplicated layers. The optimized implementation that re-uses earlier layers is in our custom implementation or in our gpt-fast implementation.
Benchmark
If you would like to measure the speedup between self-speculative decoding and autoregressive decoding, we have written this script:
from transformers import LlamaForCausalLM, LlamaConfig, LlamaTokenizer, GenerationConfig
import torch
from copy import deepcopy
from time import time
from tqdm import tqdm
prompt = "typing import List\ndef bucket_sort(A: List):"
checkpoint = "facebook/layerskip-llama2-7B"
early_exit = 4
device = "cuda" if torch.cuda.is_available() else "cpu"
max_new_tokens = 512
do_sample = True
top_p = 0.9
temperature = 0.6
warmup = 2
repeat = 10
config = LlamaConfig.from_pretrained(checkpoint)
model = LlamaForCausalLM.from_pretrained(checkpoint, config=config, torch_dtype=torch.float16)
# Draft model
# Clone main model with shared weights
weights_memo = {id(w): w for w in model.parameters()}
assistant_model = deepcopy(model, memo=weights_memo)
# Create early exit version
assistant_model.model.layers = assistant_model.model.layers[:early_exit]
del assistant_model.model.layers[early_exit:]
model.to(device)
assistant_model.to(device)
tokenizer = LlamaTokenizer.from_pretrained(checkpoint, use_fast=False)
inputs = tokenizer(prompt, return_tensors="pt").to(device)
generation_config = {
"max_new_tokens": max_new_tokens,
"do_sample": do_sample,
"top_p": top_p,
"temperature": temperature,
"pad_token_id": tokenizer.eos_token_id,
}
# Warmup
print("Warmup")
for i in tqdm(range(warmup)):
_ = model.generate(**inputs, **generation_config)
_ = model.generate(**inputs, **generation_config, assistant_model=assistant_model)
print("Autoregressive Decoding")
total_time = 0
total_tokens = 0
for i in tqdm(range(repeat)):
start = time()
outputs = model.generate(**inputs, **generation_config)
total_time += time() - start
total_tokens += outputs.numel()
print(tokenizer.batch_decode(outputs, skip_special_tokens=True)[0])
print("\n\t=========================")
print(f"\tAverage Generation Time: {total_time / repeat:.2f} s")
print(f"\tAverage Tokens per Second: {total_tokens / total_time:.2f} tokens per sec\n\n")
print("Self-Speculative Decoding")
total_time = 0
total_tokens = 0
for i in tqdm(range(repeat)):
start = time()
outputs = model.generate(**inputs, **generation_config, assistant_model=assistant_model)
total_time += time() - start
total_tokens += outputs.numel()
print(tokenizer.batch_decode(outputs, skip_special_tokens=True)[0])
print("\n\t=========================")
print(f"\tAverage Generation Time: {total_time / repeat:.2f} s")
print(f"\tAverage Tokens per Second: {total_tokens / total_time:.2f} tokens per sec\n\n")
Running this script on a single A100 NVIDIA GPU with transformers==4.34.1, torch==2.2.1, triton==2.2.0, we obtain:
Autoregressive Decoding
=========================
Average Generation Time: 12.60 s
Average Tokens per Second: 34.87 tokens per sec
Self-Speculative Decoding
=========================
Average Generation Time: 7.38 s
Average Tokens per Second: 56.10 tokens per sec
LayerSkip Codebase
Our self-speculative decoding implementation at github.com/facebookresearch/LayerSkip has an optimized version that does not consume extra memory and re-uses the weights and KV cache of earlier layers in both draft and verification stages. To run:
> git clone [email protected]:facebookresearch/LayerSkip.git
> cd LayerSkip
> conda create --name layer_skip python=3.10
> conda activate layer_skip
> pip install -r requirements.txt
> torchrun generate.py --model facebook/layerskip-llama2-7B --generation_strategy self_speculative --exit_layer 6 --num_speculations 4
You can find more details in the GitHub repo for more options and scripts.
gpt-fast
We have also implemented self-speculative decoding as a separatae branch in PyTorch's gpt-fast if you would to stack our solution on top of other optimizations like torch.compile() and quantization. Our gpt-fast implementation is optimized as it does not consume extra memory and re-uses the weights and KV cache of earlier layers in both draft and verification stages.
To run:
> git clone [email protected]:pytorch-labs/gpt-fast.git -b LayerSkip
> cd gpt-fast
> conda create --name gpt_fast python=3.10
> conda activate gpt_fast
> # Install PyTorch (check [here](https://pytorch.org/get-started/locally/) for other hardwares and operating systems)
> pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121
> pip install sentencepiece huggingface_hub tiktoken
> mkdir checkpoints
> MODEL_REPO=facebook/layerskip-llama2-7B
> ./scripts/prepare.sh $MODEL_REPO
> python generate.py --compile --checkpoint_path checkpoints/$MODEL_REPO/model.pth --top_k 100 --temperature 0.6 --self_speculative --early_exit 5 --speculate_k 3
Benchmark
- Autoregressive decoding:
> python generate.py --compile --checkpoint_path checkpoints/$MODEL_REPO/model.pth --top_k 100 --temperature 0.6
==========
Average tokens/sec: 110.50
Memory used: 13.88 GB
- Self-speculative decoding:
> python generate.py --compile --checkpoint_path checkpoints/$MODEL_REPO/model.pth --top_k 100 --temperature 0.6 --self_speculative --early_exit 5 --speculate_k 3
==========
{'tokens_per_sec': [120.16508373150057, 141.77910376715855, 132.42363092761354, 138.73840444421148, 121.55019835742718], 'accept_counts': [[32, 15, 19, 20], [50, 23, 21, 10], [31, 22, 16, 19], [41, 19, 19, 16], [35, 20, 15, 20], [47, 32, 9, 16]]}
Acceptance probs: [0.41622574955908287, 0.2310405643738977, 0.1746031746031746, 0.1781305114638448]
Mean Accepted: 1.1146384479717812
Average tokens/sec: 130.93
Memory used: 13.91 GB
Training
Our training implementation is work-in-progress. You can check this pull request for details and discussions.
Evaluation
We have provided evaluation results on various natural language and codinng tasks in the Model Card. You can view them on the top right hand-side bar on the screen. The numbers reported in this Model Card were evaluated using Eluether Evaluation Harness and BigCode Evaluation Harness, while the numbers provided in our paper were evaluated using Meta's internal codebase.
Issues
Please report any software "bug", or other problems with the models through one of the following means:
- Reporting issues with the model: https://github.com/facebookresearch/LayerSkip/issues
- Reporting risky content generated by the model: developers.facebook.com/llama_output_feedback
- Reporting bugs and security concerns: facebook.com/whitehat/info
License
See the LICENSE file.