A framework for evaluating language models
This project provides a unified framework to test generative language models on a large number of different evaluation tasks.
Features:
The Language Model Evaluation Harness is the backend for 🤗 Hugging Face's popular Open LLM Leaderboard, has been used in hundreds of papers is used internally by dozens of companies including NVIDIA, Cohere, Nous Research, Booz Allen Hamilton, and Mosaic ML.
To install the lm-eval
package from the github repository, run:
git clone https://github.com/EleutherAI/lm-evaluation-harness
cd lm-evaluation-harness
pip install -e .
We also provide a number of optional dependencies for . Extras can be installed via pip install -e ".[NAME]"
Name | Use |
---|---|
anthropic | For using Anthropic's models |
dev | You probably don't want to use this |
gptq | For loading models with GPTQ |
testing | You probably don't want to use this |
multilingual | For multilingual tokenizers |
openai | For using OpenAI's models |
promptsource | For using PromtSource prompts |
sentencepiece | For using the sentencepiece tokenizer |
vllm | For loading models with vLLM |
all | Loads all extras |
transformers
To evaluate a model hosted on the HuggingFace Hub (e.g. GPT-J-6B) on hellaswag
you can use the following command:
lm_eval --model hf \
--model_args pretrained=EleutherAI/gpt-j-6B \
--tasks hellaswag \
--device cuda:0 \
--batch_size 8
Additional arguments can be provided to the model constructor using the --model_args
flag. Most notably, this supports the common practice of using the revisions
feature on the Hub to store partially trained checkpoints, or to specify the datatype for running a model:
lm_eval --model hf \
--model_args pretrained=EleutherAI/pythia-160m,revision=step100000,dtype="float" \
--tasks lambada_openai,hellaswag \
--device cuda:0 \
--batch_size 8
Models that are loaded via both transformers.AutoModelForCausalLM
(autoregressive, decoder-only GPT style models) and transformers.AutoModelForSeq2SeqLM
(such as encoder-decoder models like T5) in Huggingface are supporteded.
Batch size selection can be automated by setting the --batch_size
flag to auto
. This will perform automatic detection of the largest batch size that will fit on your device. On tasks where there is a large difference between the longest and shortest example, it can be helpful to periodically recompute the largest batch size, to gain a further speedup. To do this, append :N
to above flag to automatically recompute the largest batch size N
times. For example, to recompute the batch size 4 times, the command would be:
lm_eval --model hf \
--model_args pretrained=EleutherAI/pythia-160m,revision=step100000,dtype="float" \
--tasks lambada_openai,hellaswag \
--device cuda:0 \
--batch_size auto:4
Alternatively, you can use lm-eval
instead of lm_eval
.
[!Note] Just like you can provide a local path to
transformers.AutoModel
, you can also provide a local path tolm_eval
via--model_args pretrained=/path/to/model
accelerate
To parallelize evaluation of HuggingFace models across multiple GPUs, we leverage the accelerate 🚀 library as follows:
accelerate launch -m lm_eval --model hf \
--tasks lambada_openai,arc_easy \
--batch_size 16
This will perform data-parallel evaluation: that is, placing a single full copy of your model onto each available GPU and splitting batches across GPUs to evaluate on K GPUs K times faster than on one.
If your model is is too large to be run on a single one of your GPUs then you can use accelerate
with Fully Sharded Data Parallel (FSDP) that splits the weights of the model across your data parallel ranks. To enable this, ensure you select YES
when asked Do you want to use FullyShardedDataParallel?
when running accelerate config
. To enable memory-efficient loading, select YES
when asked Do you want each individually wrapped FSDP unit to broadcast module parameters from rank 0 at the start?
. This will ensure only the rank 0 process loads the model and then broadcasts the parameters to the other ranks instead of having each rank load all parameters which can lead to large RAM usage spikes around the start of the script that may cause errors.
To pass even more advanced keyword arguments to accelerate
, we allow for the following arguments as well:
device_map_option
: How to split model weights across available GPUs. defaults to "auto".max_memory_per_gpu
: the max GPU memory to use per GPU in loading the model.max_cpu_memory
: the max amount of CPU memory to use when offloading the model weights to RAM.offload_folder
: a folder where model weights will be offloaded to disk if needed.To use accelerate
with the lm-eval
command, use
accelerate launch --no_python lm-eval --model ...
We also support vLLM for faster inference on supported model types.
lm_eval --model vllm \
--model_args pretrained={model_name},tensor_parallel_size={number of GPUs to use},dtype=auto,gpu_memory_utilization=0.8 \
--tasks lambada_openai \
--batch_size auto
For a full list of supported vLLM configurations, please reference our vLLM integration and the vLLM documentation.
Our library also supports the evaluation of models served via several commercial APIs, and we hope to implement support for the most commonly used performant local/self-hosted inference servers.
To call a hosted model, use:
export OPENAI_API_SECRET_KEY=YOUR_KEY_HERE
lm_eval --model openai-completions \
--model_args engine=davinci \
--tasks lambada_openai,hellaswag
Note that for externally hosted models, configs such as --device
and --batch_size
should not be used and do not function. Just like you can use --model_args
to pass arbitrary arguments to the model constructor for local models, you can use it to pass arbitrary arguments to the model API for hosted models. See the documentation of the hosting service for information on what arguments they support.
API or Inference Server | Implemented? | --model <xxx> name |
Models supported: | Request Types: |
---|---|---|---|---|
OpenAI Completions | :heavy_check_mark: | openai , openai-completions , gooseai |
up to code-davinci-002 |
generate_until , loglikelihood , loglikelihood_rolling |
OpenAI ChatCompletions | :x: Not yet - needs testing! | N/A | All ChatCompletions API models | generate_until (no logprobs) |
Anthropic | :heavy_check_mark: | anthropic |
Supported Anthropic Engines | generate_until (no logprobs) |
GooseAI | :heavy_check_mark: (not separately maintained) | openai , openai-completions , gooseai (same interface as OpenAI Completions) |
generate_until , loglikelihood , loglikelihood_rolling |
|
Textsynth | :heavy_check_mark: | textsynth |
All supported engines | generate_until , loglikelihood , loglikelihood_rolling |
Cohere | :hourglass: - blocked on Cohere API bug | N/A | All cohere.generate() engines |
generate_until , loglikelihood , loglikelihood_rolling |
Llama.cpp (via llama-cpp-python) | :heavy_check_mark: | gguf , ggml |
All models supported by llama.cpp | generate_until , loglikelihood , loglikelihood_rolling |
vLLM | :heavy_check_mark: | vllm |
Most HF Causal Language Models | generate_until , loglikelihood , loglikelihood_rolling |
Your inference server here! | ... | ... | ... | ... |
It is on our roadmap to create task variants designed to enable models which do not serve logprobs/loglikelihoods to be compared with generation performance of open-source models.
A number of other libraries contain scripts for calling the eval harness through their library. These include GPT-NeoX, Megatron-DeepSpeed, and mesh-transformer-jax.
If you have a Metal compatible Mac, you can run the eval harness using the MPS back-end by replacing --device cuda:0
with --device mps
(requires PyTorch version 2.1 or higher).
[!Note] You can inspect what the LM inputs look like by running the following command:
python write_out.py \ --tasks all_tasks \ --num_fewshot 5 \ --num_examples 10 \ --output_base_path /path/to/output/folder
This will write out one text file for each task.
To verify the data integrity of the tasks you're performing in addition to running the tasks themselves, you can use the --check_integrity
flag:
lm_eval --model openai \
--model_args engine=davinci \
--tasks lambada_openai,hellaswag \
--check_integrity
For models loaded with the HuggingFace transformers
library, any arguments provided via --model_args
get passed to the relevant constructor directly. This means that anything you can do with AutoModel
can be done with our library. For example, you can pass a local path via pretrained=
or use models finetuned with PEFT by taking the call you would run to evaluate the base model and add ,peft=PATH
to the model_args
argument:
lm_eval --model hf \
--model_args pretrained=EleutherAI/gpt-j-6b,parallelize=True,load_in_4bit=True,peft=nomic-ai/gpt4all-j-lora \
--tasks openbookqa,arc_easy,winogrande,hellaswag,arc_challenge,piqa,boolq \
--device cuda:0
GPTQ quantized models can be loaded by specifying their file names in ,gptq=NAME
(or ,gptq=True
for default names) in the model_args
argument:
lm_eval --model hf \
--model_args pretrained=model-name-or-path,gptq=model.safetensors,gptq_use_triton=True \
--tasks hellaswag
We support wildcards in task names, for example you can run all of the machine-translated lambada tasks via --task lambada_openai_mt_*
.
To save evaluation results provide an --output_path
. We also support logging model responses with the --log_samples
flag for post-hoc analysis.
Additionally, one can provide a directory with --use_cache
to cache the results of prior runs. This allows you to avoid repeated execution of the same (model, task) pairs for re-scoring.
For a full list of supported arguments, check out the interface guide in our documentation!
For more information on the library and how everything fits together, check out all of our documentation pages! We plan to post a larger roadmap of desired + planned library improvements soon, with more information on how contributors can help.
You can also ask for help, or discuss new features with the maintainers in the #lm-thunderdome channel of the EleutherAI discord! If you've used the library and have had a positive (or negative) experience, we'd love to hear from you!
To implement a new task in the eval harness, see this guide.
In general, we following the following priority list for addressing concerns about prompting and other eval details:
These are guidelines and not rules, and can be overruled in special circumstances.
We try to prioritize agreement with the procedures used by other groups to decrease the harm when people inevitably compare runs across different papers despite our discouragement of the practice. Historically, we also prioritized the implementation from "Language Models are Few Shot Learners" as our original goal was specifically to compare results with that paper.
The best way to get support is to open an issue on this repo or join the EleutherAI discord server](https://discord.gg/eleutherai). The #lm-thunderdome
channel is dedicated to developing this project and the #release-discussion
channel is for receiving support for our releases.
@misc{eval-harness,
author = {Gao, Leo and Tow, Jonathan and Abbasi, Baber and Biderman, Stella and Black, Sid and DiPofi, Anthony and Foster, Charles and Golding, Laurence and Hsu, Jeffrey and Le Noac'h, Alain and Li, Haonan and McDonell, Kyle and Muennighoff, Niklas and Ociepa, Chris and Phang, Jason and Reynolds, Laria and Schoelkopf, Hailey and Skowron, Aviya and Sutawika, Lintang and Tang, Eric and Thite, Anish and Wang, Ben and Wang, Kevin and Zou, Andy},
title = {A framework for few-shot language model evaluation},
month = sep,
year = 2021,
publisher = {Zenodo},
version = {v0.0.1},
doi = {10.5281/zenodo.5371628},
url = {https://doi.org/10.5281/zenodo.5371628}
}