[WIP] Adding GPT-NeoX-20B (#16659)
* initial * first try * working 20B * 20B tokenizers * Docs * Import fixes for missing classes * Update docs, fixup * black formatting * isort * flake * dummy objects * documentation * Documentation yml * more docs * tweaks for tests * tokenization auto * fix neox tests * test * test * einsum * address PR feedback * Documentation * Update README.md Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com> * Update src/transformers/models/gpt_neox/__init__.py Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com> * Update src/transformers/models/gpt_neox/configuration_gpt_neox.py Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com> * Apply suggestions from code review Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com> * Remove undefined LaTeX syntax * Update to full url to avoid confusion about if that's supposed to refer to the Hub * fix auto * move tests * documentation fix * more doc fixes * test refactor * fix import * fix import * fix import * fix import * fix import * style fixes * More modeling fixes Co-authored-by: Jason Phang <zp489@gr057.hpc.nyu.edu> Co-authored-by: Stella Biderman <stellabiderman@gmail.com> Co-authored-by: Sylvain Gugger <35901082+sgugger@users.noreply.github.com>
This commit is contained in:
Jason Phang
@@ -286,6 +286,8 @@
|
||||
title: GPT-J
|
||||
- local: model_doc/gpt_neo
|
||||
title: GPT Neo
|
||||
- local: model_doc/gpt_neox
|
||||
title: GPT NeoX
|
||||
- local: model_doc/hubert
|
||||
title: Hubert
|
||||
- local: model_doc/perceiver
|
||||
|
||||
@@ -95,6 +95,7 @@ The library currently contains JAX, PyTorch and TensorFlow implementations, pret
|
||||
1. **[GPT-2](model_doc/gpt2)** (from OpenAI) released with the paper [Language Models are Unsupervised Multitask Learners](https://blog.openai.com/better-language-models/) by Alec Radford*, Jeffrey Wu*, Rewon Child, David Luan, Dario Amodei** and Ilya Sutskever**.
|
||||
1. **[GPT-J](model_doc/gptj)** (from EleutherAI) released in the repository [kingoflolz/mesh-transformer-jax](https://github.com/kingoflolz/mesh-transformer-jax/) by Ben Wang and Aran Komatsuzaki.
|
||||
1. **[GPT Neo](model_doc/gpt_neo)** (from EleutherAI) released in the repository [EleutherAI/gpt-neo](https://github.com/EleutherAI/gpt-neo) by Sid Black, Stella Biderman, Leo Gao, Phil Wang and Connor Leahy.
|
||||
1. **[GPT NeoX](model_doc/gpt_neox)** (from EleutherAI) released with the paper [GPT-NeoX-20B: An Open-Source Autoregressive Language Model](https://arxiv.org/abs/2204.06745) by Sid Black, Stella Biderman, Eric Hallahan, Quentin Anthony, Leo Gao, Laurence Golding, Horace He, Connor Leahy, Kyle McDonell, Jason Phang, Michael Pieler, USVSN Sai Prashanth, Shivanshu Purohit, Laria Reynolds, Jonathan Tow, Ben Wang, Samuel Weinbach
|
||||
1. **[Hubert](model_doc/hubert)** (from Facebook) released with the paper [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://arxiv.org/abs/2106.07447) by Wei-Ning Hsu, Benjamin Bolte, Yao-Hung Hubert Tsai, Kushal Lakhotia, Ruslan Salakhutdinov, Abdelrahman Mohamed.
|
||||
1. **[I-BERT](model_doc/ibert)** (from Berkeley) released with the paper [I-BERT: Integer-only BERT Quantization](https://arxiv.org/abs/2101.01321) by Sehoon Kim, Amir Gholami, Zhewei Yao, Michael W. Mahoney, Kurt Keutzer.
|
||||
1. **[ImageGPT](model_doc/imagegpt)** (from OpenAI) released with the paper [Generative Pretraining from Pixels](https://openai.com/blog/image-gpt/) by Mark Chen, Alec Radford, Rewon Child, Jeffrey Wu, Heewoo Jun, David Luan, Ilya Sutskever.
|
||||
@@ -215,6 +216,7 @@ Flax), PyTorch, and/or TensorFlow.
|
||||
| Funnel Transformer | ✅ | ✅ | ✅ | ✅ | ❌ |
|
||||
| GLPN | ❌ | ❌ | ✅ | ❌ | ❌ |
|
||||
| GPT Neo | ❌ | ❌ | ✅ | ❌ | ✅ |
|
||||
| GPT NeoX | ❌ | ✅ | ✅ | ❌ | ❌ |
|
||||
| GPT-J | ❌ | ❌ | ✅ | ✅ | ✅ |
|
||||
| Hubert | ❌ | ❌ | ✅ | ✅ | ❌ |
|
||||
| I-BERT | ❌ | ❌ | ✅ | ❌ | ❌ |
|
||||
|
||||
76
docs/source/en/model_doc/gpt_neox.mdx
Normal file
76
docs/source/en/model_doc/gpt_neox.mdx
Normal file
@@ -0,0 +1,76 @@
|
||||
<!--Copyright 2022 The HuggingFace Team. All rights reserved.
|
||||
|
||||
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with
|
||||
the License. You may obtain a copy of the License at
|
||||
|
||||
http://www.apache.org/licenses/LICENSE-2.0
|
||||
|
||||
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on
|
||||
an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the
|
||||
specific language governing permissions and limitations under the License.
|
||||
-->
|
||||
|
||||
# GPT-NeoX
|
||||
|
||||
## Overview
|
||||
|
||||
We introduce GPT-NeoX-20B, a 20 billion parameter autoregressive language model trained on the Pile, whose weights will
|
||||
be made freely and openly available to the public through a permissive license. It is, to the best of our knowledge,
|
||||
the largest dense autoregressive model that has publicly available weights at the time of submission. In this work,
|
||||
we describe GPT-NeoX-20B's architecture and training and evaluate its performance on a range of language-understanding,
|
||||
mathematics, and knowledge-based tasks. We find that GPT-NeoX-20B is a particularly powerful few-shot reasoner and
|
||||
gains far more in performance when evaluated five-shot than similarly sized GPT-3 and FairSeq models. We open-source
|
||||
the training and evaluation code, as well as the model weights, at [https://github.com/EleutherAI/gpt-neox](https://github.com/EleutherAI/gpt-neox).
|
||||
|
||||
Development of the model was led by Sid Black, Stella Biderman and Eric Hallahan, and the model was trained with
|
||||
generous the support of [CoreWeave](https://www.coreweave.com/).
|
||||
|
||||
GPT-NeoX-20B was trained with fp16, thus it is recommended to initialize the model as follows:
|
||||
|
||||
```python
|
||||
model = GPTNeoXForCausalLM.from_pretrained("EleutherAI/gpt-neox-20b").half().cuda()
|
||||
```
|
||||
|
||||
GPT-NeoX-20B also has a different tokenizer from the one used in GPT-J-6B and GPT-Neo. The new tokenizer allocates
|
||||
additional tokens to whitespace characters, making the model more suitable for certain tasks like code generation.
|
||||
|
||||
### Generation
|
||||
|
||||
The `generate()` method can be used to generate text using GPT Neo model.
|
||||
|
||||
```python
|
||||
>>> from transformers import GPTNeoXForCausalLM, GPTNeoXTokenizerFast
|
||||
|
||||
>>> model = GPTNeoXForCausalLM.from_pretrained("EleutherAI/gpt-neox-20b")
|
||||
>>> tokenizer = GPTNeoXTokenizerFast.from_pretrained("EleutherAI/gpt-neox-20b")
|
||||
|
||||
>>> prompt = "GPTNeoX20B is a 20B-parameter autoregressive Transformer model developed by EleutherAI."
|
||||
|
||||
>>> input_ids = tokenizer(prompt, return_tensors="pt").input_ids
|
||||
|
||||
>>> gen_tokens = model.generate(
|
||||
... input_ids,
|
||||
... do_sample=True,
|
||||
... temperature=0.9,
|
||||
... max_length=100,
|
||||
... )
|
||||
>>> gen_text = tokenizer.batch_decode(gen_tokens)[0]
|
||||
```
|
||||
|
||||
## GPTNeoXConfig
|
||||
|
||||
[[autodoc]] GPTNeoXConfig
|
||||
|
||||
## GPTNeoXTokenizerFast
|
||||
|
||||
[[autodoc]] GPTNeoXTokenizerFast
|
||||
|
||||
## GPTNeoXModel
|
||||
|
||||
[[autodoc]] GPTNeoXModel
|
||||
- forward
|
||||
|
||||
## GPTNeoXForCausalLM
|
||||
|
||||
[[autodoc]] GPTNeoXForCausalLM
|
||||
- forward
|
||||
Reference in New Issue
Block a user