Guide to using pre-trained large language models of source code

Vincent Hellendoorn

Last update: Dec 28, 2022

Related tags

Overview

Large Models of Source Code

I occasionally train and publicly release large neural language models on programs, including PolyCoder. Here, I describe how to use these.

Setup
Models (incl. PolyCoder)
Datasets
Evaluation
How to cite

Getting Started

All current models were trained using the GPT NeoX toolkit. First, download a pretrained checkpoint as described below and then use this either with a Docker image or through our fork of this toolkit from source to generate code or replicate our evaluation.

Retrieving Checkpoints

Checkpoint files for training PolyCoder are hosted on this public Zenodo repository. See this section for details on currently available models. Model checkpoints range up to 6GB, which is also the amount of GPU memory they require to run (running on CPU is neither tested nor recommended). Download and untar a checkpoint file (in this case for a 2.7B parameter model trained for 150K steps) to a directory called checkpoints/, using:

mkdir checkpoints
cd checkpoints
wget https://zenodo.org/record/6363556/files/2-7B-150K.tar
tar -xvf 2-7B-150K.tar

From Source

We maintain a public fork of the NeoX repository here, which includes the (minor) changes we made to the codebase to allow for tabs & newlines in the tokenization, and also includes instructions for running the perplexity and HumanEval tasks. Note that this repository uses a forked version of the LM Evaluation Harness with the code benchmark from our work.

Building this repository should match the process for GPT-NeoX almost exactly. You may also use the Docker image mentioned next, but mounting a checkout of the latest version of this fork over the /gpt-neox directory inside the container. Once set up generate.py entrypoint (described below) for free-form code generation, or use one of the commands here to calculate perplexity and HumanEval results as in the paper.

Via Docker

A base Docker image containing a slightly modified version of the gpt-neox repository is available via DockerHub:

docker pull vhellendoorn/code-lms-neox:base

This image can be used together with a checkpoint file hosted on this public Zenodo repository. The base Docker image size is 5.4GB. Once a checkpoint has been retrieved, start the container with the following commands (substituting another GPU device index if needed):

nvidia-docker run --rm -it -e NVIDIA_VISIBLE_DEVICES=0 --shm-size=1g --ulimit memlock=-1 --mount type=bind,src=$PWD/checkpoints,dst=/gpt-neox/checkpoints vhellendoorn/code-lms-neox:base

Code Generation

The following command can be used to generate code from a prompt:

sudo ./deepy.py generate.py configs/text_generation.yml checkpoints/configs/local_setup.yml checkpoints/configs/2-7B.yml

Note: if not using the 2.7B parameter model, replace the final config file with the appropriate model size (e.g., small = 160M parameters, medium = 405M).

Once the checkpoint has been loaded, you can feed it an example such as def return1():\n """Returns 1."""\n (note the whitespace tokens) and watch it predict return 1 (and then probably a bunch of other returnX methods, depending on the sample).

The modifications to gpt-neox mentioned above center around the need to allow tabs and newlines in the prompt input. For the interactive mode, these can be added using their escaped versions (\t, \n); when using file-based input, the project will read the entire file instead of treating each line as a prompt. By default, the command below will create an interactive prompt and return relatively short outputs (256 tokens) with a sampling temperature of 0.5; this behavior can be changed in /gpt-neox/checkpoints/configs/text_generation.yml.

A lower temperature (e.g., 0.2) will produce more consistent and plausible (to the model) predictions; a higher temperature such as the default may be useful for generating and evaluating many candidates (see our paper for recommendations). For the latter setting, consider switching to the input-file mode and providing an entire snippet (without escaping whitespace) in the corresponding file

Multi-lingual Models

Several models have been trained on a large corpus of code spanning 12 programming languages. This includes a 2.7B parameter model (nick-named PolyCoder, trained for 100K and 150K steps), a 405M parameter model (100K & 150K steps) and a 160M parameter model (150K steps).

Available Models

All models are available at a public Zenodo repository, in the form of .tar files with fairly self-explanatory names (e.g., 2-7B-100K => a 2.7B parameter model trained for 100K steps). Currently available models include:

GPT2 - 2.7B: A 32 layer, 2,560 dimensional Transformer model, trained with a batch size of 128 sequences (256K tokens). Models available both at 100K and at 150K steps steps.
- Note that GPT-Neox' default config for this model was modified to reduce the number of training steps (and learning rate decay steps accordingly) to 160K, down from 320K, to better match the available training resources. Hence, this model may not have reached its peak performance.
GPT2 - 0.4B: A 24 layer, 1,024 dimensional Transformer model based on the medium config, trained with 256K tokens per batch.
GPT2 - 160M: A 12 layer, 768 dimensional Transformer model based on the small config, trained with 256K tokens per batch.

Training Process

Training was done on 4 to 8 NVIDIA RTX 8000 GPUs, largely following the standard config values, except also enabling "scaled-upper-triang-masked-softmax-fusion" and "bias-gelu-fusion" for performance and slightly changing the batch size (see model details), data split (changed to 98.9%, 0.1%, 1%), initial loss scale (2^16), and print/eval intervals.

The below image shows the loss curve of the various models' training process in terms of validation loss.

Caveats

The trained models come with a few minor known limitations:

This model was not trained to solve programming problems and may not perform well on a benchmark such as HumanEval. Models like Codex (powering Copilot) are pretrained on natural language, which may boost their ability to interpret NL prompts; this model only learned language from comments in code.
The model appears to start generating a random new file once it reaches the (predicted) end of the current one. It is possible that the end-of-document token was not properly added to the training data.
Whitespace is very important to the model, since no preprocessing was done on the input files. For instance, the following snippet will yield poor predictions, because in Java we would never expect an instance-method at the top-level, as is indicated by the single level of (\t) indentation of the two lines within this method:

public int getTotalWeight(List<Integer> weights) {\n\t// Sum weights in parallel.\n\treturn

Adjusting the indentation makes it predict more reasonable continuations:

public int getTotalWeight(List<Integer> weights) {\n\t\t// Sum weights in parallel.\n\t\treturn

The Codex model discusses controlling for this to increase usability; this may be worth doing in a future version of the model.

Datasets

249GB Multi-Lingual Corpus

This is the corpus used to train PolyCoder.

The datasets were cloned overnight on October 9-10, 2021. To mine a similar training set, see Data.

The list of file paths can be downloaded from: https://zenodo.org/record/6363556/files/index.zip. Each row in the file is the file path along with its SHA-256 hash, to ease deduplication. That is, the hashes allow checking if files from any future test set were already contained in the training set.

The data collection and filtering process is described in detail in the paper and below. The final, filtered dataset statistics are:

Language	Repositories	Size(GB)	Files
C	10,749	55G	3,037,112
C#	9,511	21G	2,514,494
C++	13,726	52G	4,289,506
Go	12,371	15G	1,416,789
Java	15,044	41G	5,120,129
JavaScript	25,144	22G	1,774,174
PHP	9,960	13G	1,714,058
Python	25,446	16G	1,550,208
Ruby	5,826	4.1G	674,343
Rust	4,991	3.5G	304,842
Scala	1,497	1.8G	245,100
TypeScript	12,830	9.2G	1,441,926

Data Collection & Filtering

I cloned the most popular repositories for 12 popular programming languages with at least 50 stars (stopping at ~25K per langauge) from GitHub in October 2021. For each project, each file belonging to the majority-language of that project was extracted, yielding the training set below (after cleaning). This initial, unfiltered dataset spanned 631GB and 38.9M files.

Next, similar to Codex and CodeParrot, very large (>1MB) and very short (<100 tokens) files were filtered out, reducing the dataset to 424GB. Files were then deduplicated based on a hash of their content, which reduced the number of files by another 30% or so, leaving 249GB of data and 24.1M files. No tokenization filters were applied; the model processes entire files including all comments. A code-specific vocabulary was constructed on a random 5% subset of the files above.

Evaluation

Please find detailed instructions for replicating our perplexity and HumanEval results on our public fork of the NeoX repository. This in turn leverages our extension of the LM Evaluation Harness.

Evaluating Codex

To download the test sets that we used in the paper (12 programming languages), use:

wget https://zenodo.org/record/6363556/files/unseen_test_sets.tar.gz
tar -xvzf unseen_test_sets.tar.gz

To get perplexity results on these samples using Codex' API, use:

export OPENAI_API_KEY=<YOUR OPEN AI API KEY>
python3 -u Evaluation/eval_codex_all.py --dirs Code-sampled100

Where <YOUR OPEN AI API KEY> is a private string that can be obtained by signing up for OpenAI's beta.

As of March 2022, getting an API Key is free for 3 months, and afterwards a credit card needs to be entered. However, even after entering a credit card, using our evaluation script does not lead to any costs.

Results - HumanEval

These are PolyCoder's results on the HumanEval benchmark:

Model	Pass@1	Pass@10	Pass@100
PolyCoder (160M)	2.13%	3.35%	4.88%
PolyCoder (400M)	2.96%	5.29%	11.59%
PolyCoder (2.7B)	5.59%	9.87%	17.68%
CodeParrot (110M)	3.80%	6.57%	12.78%
CodeParrot (1.5B)	3.58%	8.03%	14.96%
GPT-Neo (125M)	0.75%	1.88%	2.97%
GPT-Neo (1.3B)	4.79%	7.47%	16.30%
GPT-Neo (2.7B)	6.41%	11.27%	21.37%
GPT-J (6B)	11.62%	15.74%	27.74%
Codex (300M)	13.17%	20.37%	36.27%
Codex (2.5B)	21.36%	35.42%	59.50%
Codex (12B)	28.81%	46.81%	72.31%

Results - Multilingual Language Modeling

These are the perplexity results of PolyCoder on the multilingual test sets:

Language	Perplexity
C	2.3464
C#	2.5832
C++	2.9189
Go	2.567
Java	2.9194
JavaScript	3.0611
PHP	3.6954
Python	3.1767
Ruby	3.9742
Rust	3.2449
Scala	3.8735
TypeScript	3.6143

A comparison with the other models is available in Figure 6 in the paper:

Citation

A Systematic Evaluation of Large Language Models of Code

@article{xu2022systematic,
  title={A Systematic Evaluation of Large Language Models of Code},
  author={Xu, Frank F and Alon, Uri and Neubig, Graham and Hellendoorn, Vincent J},
  journal={arXiv preprint arXiv:2202.13169},
  year={2022}
}

Comments

Convert GPT-NeoX to HuggingFace

This PR includes a script named convert_neox_pt_to_huggingface_neox.py, which is used to convert a PolyCoder checkpoint trained by GPT-NeoX into HuggingFace format. A transformers.GPTNeoXConfig file that matches the checkpoint is also provided.

I have checked with a 0.4B model, which is trained following the medium.yml config. The greedy-decoding outputs by GPT-NeoX's inference script and HF's generate() are identical.

NOTE: The HuggingFace model type I used is "gpt_neox", but the model architecture needs to be adjusted. Please refer to this PR.

opened by NinedayWang 9
Docker Image code generation fails

Hello!

While trying to run polycoder in a dockerized setup, we bumped into the error: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling cublasGemmEx( handle, opa, opb, m, n, k, &falpha, a, CUDA_R_16F, lda, b, CUDA_R_16F, ldb, &fbeta, c, CUDA_R_16F, ldc, CUDA_R_32F, CUBLAS_GEMM_DFALT_TENSOR_OP)

Could you help us to get over this problem?

This is odd, because I guess the docker solution should have been error free.

We're trying to run the 160M version on a Ubuntu 22.04 machine with a GTX 1060 6GB by using the provided command to start the container.

Here is the full log: https://pastebin.com/JzayrXUr

opened by Longman-Stan 9

Code generation fails if more than 1 token in input

Hi, I encounter quite the same problem as in #23 but in a slightly different GPU context. As him/her, I can generate for a single token input but it fails as soon as I add a character to the input.

I had to use singularity instead of docker as I had no root access on this cluster. For this to work, I had to copy the /gpt-neox out of the image and bind-mount it in order to have it in read-write mode.

I start it like that:

srun --account=xxx --partition=xxx --gres=gpu:1 --time=0-2:00:00 --mem=70G --pty  bash
[xxx@nodexx xxx]$ singularity shell --nv -B $PWD/gpt-neox:/gpt-neox -B $PWD/checkpoints-0-4B:/gpt-neox/checkpoints --writable-tmpfs code-lms-neox_base.sif 
./deepy.py generate.py configs/text_generation.yml checkpoints/configs/local_setup.yml checkpoints/configs/medium.yml

The full output including the correct generation for for and the failure for fora is here: https://paste.libre-service.eu/?f9efce75b3836b4b#BNZq675ihp2LnSqRywMa2Y9EtwzCnu4tZqUwt9o57Q9L

The GPU is a Quadro P5000. Here is the output of nvidia-smi:

nvidia-smi
Wed Jun 29 15:14:34 2022
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.57.02    Driver Version: 470.57.02    CUDA Version: 11.4     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|                               |                      |               MIG M. |
|===============================+======================+======================|
|   0  Quadro P5000        On   | 00000000:1B:00.0 Off |                  Off |
| 52%   80C    P0   163W / 180W |   3133MiB / 16278MiB |    100%      Default |
|                               |                      |                  N/A |

The GPU memory is never above 2GB (on 16 available).

I have no more ideas to try to solve this problem.

(Note that I first tried to install from source but it failed when installing the requirements)

opened by kleag 5

Hugging Face's Hub

Hello! First I would like to congratulate you on the fantastic work done with PolyCoder and the paper.

Second, I wanted to know if there are plans to release PolyCoder in Hugging Face any time soon? I saw in #13 that in March something about that possibility was commented, but couldn't find it in the hub.

There is still complicated to run the model and adapt it to new scenarios, and making it available by Hugging Face API would facilitate everything.

Thanks!

opened by Otavio-Parraga 3
How to evaluate the perplexity of PolyCoder/CodeParrot?

Hello,

Thank you for the awesome PolyCoder project, I really think it helps a lot for evaluating all the Code PTMs. However, I only found a script to evaluate the perplexity of codex, it's hard to reproduce the perplexity benchmark results when compared among codex, Polycoder and other PTMs. Is it possible that you guys update a polycoder/codeparrot perplexity evalution script?

Thanks again, Sen

opened by nforest 3

PlaidML support for (MacOS)

I am currently on a Mac Pro with a AMD Radeon Pro 580X but the code generation doesn't complete due to this error.

NeoXArgs.from_ymls() ['configs/text_generation.yml', 'configs/local_setup.yml', 'configs/2-7B.yml']
INFO:root:NeoXArgs.calculate_derived() Total number of GPUs determined to be: 0
Traceback (most recent call last):
  File "/Users/xoxrumblelorexox/Desktop/Workbash/polyai/checkpoints/checkpoints/./gpt-neox-main/deepy.py", line 28, in <module>
    neox_args = NeoXArgs.consume_deepy_args()
  File "/Users/xoxrumblelorexox/Desktop/Workbash/polyai/checkpoints/checkpoints/gpt-neox-main/megatron/neox_arguments/arguments.py", line 321, in consume_deepy_args
    neox_args = cls.from_ymls(
  File "/Users/xoxrumblelorexox/Desktop/Workbash/polyai/checkpoints/checkpoints/gpt-neox-main/megatron/neox_arguments/arguments.py", line 201, in from_ymls
    return cls(**config)
  File "<string>", line 186, in __init__
  File "/Users/xoxrumblelorexox/Desktop/Workbash/polyai/checkpoints/checkpoints/gpt-neox-main/megatron/neox_arguments/arguments.py", line 106, in __post_init__
    self.calculate_derived()
  File "/Users/xoxrumblelorexox/Desktop/Workbash/polyai/checkpoints/checkpoints/gpt-neox-main/megatron/neox_arguments/arguments.py", line 675, in calculate_derived
    self.check_batch_parameters(
  File "/Users/xoxrumblelorexox/Desktop/Workbash/polyai/checkpoints/checkpoints/gpt-neox-main/megatron/neox_arguments/arguments.py", line 589, in check_batch_parameters
    assert (
AssertionError: Train batch size: 0 has to be greater than 0

After a while i realised that tensorflow doesn't recognise non-Nividia GPUs. There should be a disclaimer about this on the page. The only way around i found was to use PlaidML, but i am still getting the same error (probably because PlaidML only supports Keras)

Sorry, if already known.

opened by XoxRumbleLorexoX 2

Dataset index.txt file contains some corrupted entries
Hi,

Thanks for open sourcing the dataset and PolyCoder.

I'm looking at the dataset: https://zenodo.org/record/6363556/files/index.zip , From the README, it seems that each line in the index.txt should be in the form of {language}__{organization}__{project}__{full__file__path}\tSHA, however after parsing there are few lines that seems to be malformed.

I've attached the malformed entries below

Line 2818009 : Upside Down Numbers Line 2818010 : Upside Down Numbers Line 2818011 : Upside Down Numbers Line 2818012 : Upside Down Numbers Line 2818013 : Upside Down Numbers__main.cpp 0070bf300d9f1bf6ec6533142fbbaa4de8ff65374da8d29e6a85cba5d0ad38df Line 2818158 : Phone Number Combinations__main.cpp 39ae7219b6377846a3792efcf6db5a9cc1b949652a3cc76edbe5c368f37b90a1 Line 11100410 : .java 3f49f41560cd7a5ea7c2d31120d98dfd2f56da204b6286e099af69d629ba3041```
opened by brutalsavage 2
cuda error

hi,i run 2-7B model ,this is my command :sudo ./deepy.py generate.py configs/text_generation.yml checkpoints/configs/local_setup.yml checkpoints/configs/2-7B.yml，and I get this error message. [2022-03-28 14:07:59,567] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=0 file=checkpoints/global_step150000/layer_00-model_00-model_states.pt

would help me solve this problem!! thanks.

opened by liuyongqiangjava 2
运行问题

您好可以帮我看下报错吗 No CUDA runtime is found, using CUDA_HOME='/usr/local/cuda' NeoXArgs.from_ymls() ['configs/text_generation.yml', 'checkpoints/configs/local_setup.yml', 'checkpoints/configs/2-7B.yml'] INFO:root:NeoXArgs.calculate_derived() Total number of GPUs determined to be: 0 Traceback (most recent call last): File "./deepy.py", line 29, in <module> neox_args = NeoXArgs.consume_deepy_args() File "/gpt-neox/megatron/neox_arguments/arguments.py", line 304, in consume_deepy_args neox_args = cls.from_ymls(paths_to_yml_files=conf_files, overwrite_values=overwrite_values) File "/gpt-neox/megatron/neox_arguments/arguments.py", line 199, in from_ymls return cls(**config) File "<string>", line 184, in __init__ File "/gpt-neox/megatron/neox_arguments/arguments.py", line 106, in __post_init__ self.calculate_derived() File "/gpt-neox/megatron/neox_arguments/arguments.py", line 656, in calculate_derived self.check_batch_parameters( File "/gpt-neox/megatron/neox_arguments/arguments.py", line 570, in check_batch_parameters assert ( AssertionError: Train batch size: 0 has to be greater than 0 谢谢～

opened by ipsaxiaobo 2
Code Completion Support

Hi, thanks for your nice trained model~ I would like to add it to a code completion(not code generation) backend, but I cannot find an API to do this, such as using gpt cache to decode one step and go through the whole code files to calculate top-1 accuracy.

opened by HoratioJSY 2
Plan on Releasing Generated Sample C Code

Hi there, nice job on this work! I'm wondering are you planning on releasing some generated sample C code? Just be curious about what they look like. For judging the functional correctness of the generated Python code, I know you used HumanEval to evaluate that, did you conduct a similar functionality check on C code also? Thanks much!

opened by Lightninghkm 2

Context Prompt - RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling ```cublasGemmEx( handle, opa, opb, m, n, k, &falpha, a, CUDA_R_16F, lda, b, CUDA_R_16F, ldb, &fbeta, c, CUDA_R_16F, ldc, CUDA_R_32F, CUBLAS_GEMM_DFALT_TENSOR_OP)```

I am trying to setup Polycoder inferencing on my machine with 2xP100 GPUs, and use the docker command as available in README:

nvidia-docker run --rm -it -e NVIDIA_VISIBLE_DEVICES=0,1 --shm-size=1g --ulimit memlock=-1 --mount type=bind,src=$PWD/Downloads/checkpoints/checkpoints-2-7B,dst=/gpt-neox/checkpoints vhellendoorn/code-lms-neox:base

And then within the container:

 sudo ./deepy.py generate.py configs/text_generation.yml checkpoints/configs/local_setup.yml checkpoints/configs/2-7B.yml

The following is the output (stdout+stderr):

NeoXArgs.from_ymls() ['configs/text_generation.yml', 'checkpoints/configs/local_setup.yml', 'checkpoints/configs/2-7B.yml']
INFO:root:NeoXArgs.calculate_derived() Total number of GPUs determined to be: 2
-------------------- arguments --------------------
  attention_config ................ ['global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global', 'global']updated
  attention_dropout ............... 0...........................updated
  batch_size ...................... 8...........................updated
  bias_gelu_fusion ................ True........................updated
  checkpoint_activations .......... True........................updated
  clip_grad ....................... 1.0.........................updated
  config_files .................... {'text_generation.yml': '# Parameters used for text generation\n# Make sure `load` is specified somewhere else\n{\n  # Text gen type: `input-file`, `unconditional` or `interactive`\n  "text-gen-type": "interactive",\n \n  # Params for all\n  "maximum_tokens": 256,\n  "temperature": 0.5,\n  "top_p": 0.0,\n  "top_k": 0,\n  "recompute": false,\n  \n  # `unconditional`: samples\n  "num-samples": 10,\n\n  # input/output file\n  "sample-input-file": "sample_input.txt",\n  "sample-output-file": "sample_output.txt",\n}', 'local_setup.yml': '# Suggested data paths when using GPT-NeoX locally\n{\n  "data-path": "data/code/code_text_document",\n  \n  # or for weighted datasets: \n  # "train-data-paths": ["data/enron/enron_text_document", "data/enron/enron_text_document"],\n  # "test-data-paths": ["data/enron/enron_text_document", "data/enron/enron_text_document"],\n  # "valid-data-paths": ["data/enron/enron_text_document", "data/enron/enron_text_document"],\n  # "train-data-weights": [1., 2.],\n  # "test-data-weights": [2., 1.],\n  # "valid-data-weights": [0.5, 0.4],\n\n  # If weight_by_num_documents is True, Builds dataset weights from a multinomial distribution over groups of data according to the number of documents in each group. \n  # WARNING: setting this to True will override any user provided weights\n  # "weight_by_num_documents": false,\n  # "weighted_sampler_alpha": 0.3,\n\n  "vocab-file": "data/code-vocab.json",\n  "merge-file": "data/code-merges.txt",\n\n  "save": "checkpoints",\n  "load": "checkpoints",\n  "checkpoint_validation_with_forward_pass": False,\n  \n  "tensorboard-dir": "tensorboard",\n  "log-dir": "logs",\n  "use_wandb": True,\n  "wandb_host": "https://api.wandb.ai",\n  "wandb_project": "neox"\n}', '2-7B.yml': '# GPT-2 pretraining setup\n{\n   # parallelism settings ( you will want to change these based on your cluster setup, ideally scheduling pipeline stages\n   # across the node boundaries )\n   "pipe-parallel-size": 1,\n   "model-parallel-size": 1,\n\n   # model settings\n   "num-layers": 32,\n   "hidden-size": 2560,\n   "num-attention-heads": 32,\n   "seq-length": 2048,\n   "max-position-embeddings": 2048,\n   "norm": "layernorm",\n   "pos-emb": "rotary",\n   "no-weight-tying": true,\n\n   # these should provide some speedup but takes a while to build, set to true if desired\n   "scaled-upper-triang-masked-softmax-fusion": true,\n   "bias-gelu-fusion": true,\n\n   # optimizer settings\n   "zero_allow_untested_optimizer": true,\n   "optimizer": {\n     "type": "adam",\n     "params": {\n       "lr": 0.00016,\n       "betas": [0.9, 0.999],\n       "eps": 1.0e-8,\n     }\n   },\n   "zero_optimization": {\n    "stage": 1,\n    "allgather_partitions": True,\n    "allgather_bucket_size": 500000000,\n    "overlap_comm": True,\n    "reduce_scatter": True,\n    "reduce_bucket_size": 500000000,\n    "contiguous_gradients": True,\n    "cpu_offload": False\n  },\n\n   # batch / data settings\n   "train_micro_batch_size_per_gpu": 8,\n   "gradient_accumulation_steps": 4,\n   "data-impl": "mmap",\n   "split": "989,10,1",\n\n   # activation checkpointing\n   "checkpoint-activations": true,\n   "checkpoint-num-layers": 1,\n   "partition-activations": true,\n   "synchronize-each-layer": true,\n\n   # regularization\n   "gradient_clipping": 1.0,\n   "weight-decay": 0,\n   "hidden-dropout": 0,\n   "attention-dropout": 0,\n\n   # precision settings\n   "fp16": { \n     "fp16": true,\n     "enabled": true,\n     "loss_scale": 0,\n     "initial_scale_power": 16,\n     "loss_scale_window": 1000,\n     "hysteresis": 2,\n     "min_loss_scale": 1\n   },\n\n   # misc. training settings\n   "train-iters": 160000,\n   "lr-decay-iters": 160000,\n   "distributed-backend": "nccl",\n   "lr-decay-style": "cosine",\n   "warmup": 0.01,\n   "save-interval": 1000,\n   "eval-interval": 1000,\n   "eval-iters": 10,\n\n   # logging\n   "log-interval": 100,\n   "steps_per_print": 10,\n   "keep-last-n-checkpoints": 1,\n   "wall_clock_breakdown": true,\n}\n'}updated
  data_impl ....................... mmap........................updated
  data_path ....................... data/code/code_text_documentupdated
  dynamic_loss_scale .............. True........................updated
  eval_iters ...................... 10..........................updated
  fp16 ............................ {'fp16': True, 'enabled': True, 'loss_scale': 0, 'initial_scale_power': 16, 'loss_scale_window': 1000, 'hysteresis': 2, 'min_loss_scale': 1}updated
  gas ............................. 4...........................updated
  global_num_gpus ................. 2...........................updated
  gradient_accumulation_steps ..... 4...........................updated
  gradient_clipping ............... 1.0.........................updated
  hidden_dropout .................. 0...........................updated
  hidden_size ..................... 2560........................updated
  is_pipe_parallel ................ True........................updated
  keep_last_n_checkpoints ......... 1...........................updated
  load ............................ checkpoints.................updated
  log_dir ......................... logs........................updated
  log_interval .................... 100.........................updated
  lr .............................. 0.00016.....................updated
  lr_decay_iters .................. 160000......................updated
  lr_decay_style .................. cosine......................updated
  max_position_embeddings ......... 2048........................updated
  maximum_tokens .................. 256.........................updated
  merge_file ...................... data/code-merges.txt........updated
  no_weight_tying ................. True........................updated
  num_attention_heads ............. 32..........................updated
  num_layers ...................... 32..........................updated
  num_samples ..................... 10..........................updated
  optimizer ....................... {'type': 'adam', 'params': {'lr': 0.00016, 'betas': [0.9, 0.999], 'eps': 1e-08}}updated
  partition_activations ........... True........................updated
  pipe_parallel_size .............. 1...........................updated
  pos_emb ......................... rotary......................updated
  precision ....................... fp16........................updated
  sample_input_file ............... sample_input.txt............updated
  sample_output_file .............. sample_output.txt...........updated
  save ............................ checkpoints.................updated
  save_interval ................... 1000........................updated
  scaled_upper_triang_masked_softmax_fusion  True...............updated
  seq_length ...................... 2048........................updated
  sparsity_config ................. {}..........................updated
  split ........................... 989,10,1....................updated
  synchronize_each_layer .......... True........................updated
  temperature ..................... 0.5.........................updated
  tensorboard_dir ................. tensorboard.................updated
  text_gen_type ................... interactive.................updated
  train_batch_size ................ 64..........................updated
  train_iters ..................... 160000......................updated
  train_micro_batch_size_per_gpu .. 8...........................updated
  use_wandb ....................... True........................updated
  user_script ..................... generate.py.................updated
  vocab_file ...................... data/code-vocab.json........updated
  wall_clock_breakdown ............ True........................updated
  wandb_group ..................... jtRPtjruy7PQkWHayfg7cH_6sweym4supdated
  weight_decay .................... 0...........................updated
  zero_allgather_bucket_size ...... 500000000...................updated
  zero_allow_untested_optimizer ... True........................updated
  zero_contiguous_gradients ....... True........................updated
  zero_optimization ............... {'stage': 1, 'allgather_partitions': True, 'allgather_bucket_size': 500000000, 'overlap_comm': True, 'reduce_scatter': True, 'reduce_bucket_size': 500000000, 'contiguous_gradients': True, 'cpu_offload': False}updated
  zero_reduce_bucket_size ......... 500000000...................updated
  zero_reduce_scatter ............. True........................updated
  zero_stage ...................... 1...........................updated
  activation ...................... gelu........................default
  adlr_autoresume ................. False.......................default
  adlr_autoresume_interval ........ 1000........................default
  amp ............................. None........................default
  apply_query_key_layer_scaling ... False.......................default
  attention_softmax_in_fp32 ....... False.......................default
  bias_dropout_fusion ............. False.......................default
  char_level_ppl .................. False.......................default
  checkpoint_in_cpu ............... False.......................default
  checkpoint_num_layers ........... 1...........................default
  checkpoint_validation_with_forward_pass  False................default
  contiguous_checkpointing ........ False.......................default
  deepscale ....................... False.......................default
  deepscale_config ................ None........................default
  deepspeed ....................... True........................default
  deepspeed_activation_checkpointing  True......................default
  deepspeed_mpi ................... False.......................default
  detect_nvlink_pairs ............. False.......................default
  distributed_backend ............. nccl........................default
  do_test ......................... None........................default
  do_train ........................ None........................default
  do_valid ........................ None........................default
  dump_state ...................... False.......................default
  eod_mask_loss ................... False.......................default
  eval_interval ................... 1000........................default
  eval_results_prefix ............. ............................default
  eval_tasks ...................... None........................default
  exclude ......................... None........................default
  exit_interval ................... None........................default
  finetune ........................ False.......................default
  flops_profiler .................. None........................default
  fp16_lm_cross_entropy ........... False.......................default
  fp32_allreduce .................. False.......................default
  git_hash ........................ 98683ae.....................default
  gmlp_attn_dim ................... 64..........................default
  gpt_j_residual .................. False.......................default
  gradient_noise_scale_cpu_offload  False.......................default
  gradient_noise_scale_n_batches .. 5...........................default
  gradient_predivide_factor ....... 1.0.........................default
  hostfile ........................ None........................default
  hysteresis ...................... 2...........................default
  include ......................... None........................default
  init_method ..................... normal......................default
  init_method_std ................. 0.02........................default
  iteration ....................... None........................default
  launcher ........................ pdsh........................default
  layernorm_epsilon ............... 1e-05.......................default
  lazy_mpu_init ................... False.......................default
  local_rank ...................... None........................default
  log_grad_norm ................... False.......................default
  log_gradient_noise_scale ........ False.......................default
  log_optimizer_states ............ False.......................default
  log_param_norm .................. False.......................default
  loss_scale ...................... None........................default
  loss_scale_window ............... 1000.0......................default
  make_vocab_size_divisible_by .... 128.........................default
  master_addr ..................... None........................default
  master_port ..................... 29500.......................default
  min_lr .......................... 0.0.........................default
  min_scale ....................... 1.0.........................default
  mmap_warmup ..................... False.......................default
  model_parallel_size ............. 1...........................default
  no_load_optim ................... False.......................default
  no_load_rng ..................... False.......................default
  no_save_optim ................... False.......................default
  no_save_rng ..................... False.......................default
  norm ............................ layernorm...................default
  num_gpus ........................ None........................default
  num_nodes ....................... -1..........................default
  num_unique_layers ............... None........................default
  num_workers ..................... 2...........................default
  onnx_safe ....................... False.......................default
  optimizer_type .................. adam........................default
  output_layer_init_method ........ scaled_normal...............default
  output_layer_parallelism ........ row.........................default
  override_lr_scheduler ........... False.......................default
  padded_vocab_size ............... None........................default
  param_sharing_style ............. grouped.....................default
  pipe_partition_method ........... type:transformer|mlp........default
  prescale_gradients .............. False.......................default
  profile_backward ................ False.......................default
  rank ............................ None........................default
  recompute ....................... False.......................default
  rms_norm_epsilon ................ 1e-08.......................default
  rotary_emb_base ................. 10000.......................default
  rotary_pct ...................... 1.0.........................default
  rpe_max_distance ................ 128.........................default
  rpe_num_buckets ................. 32..........................default
  scaled_masked_softmax_fusion .... False.......................default
  scalenorm_epsilon ............... 1e-08.......................default
  scheduler ....................... None........................default
  seed ............................ 1234........................default
  short_seq_prob .................. 0.1.........................default
  soft_prompt_tuning .............. None........................default
  sparse_gradients ................ False.......................default
  steps_per_print ................. 10..........................default
  test_data_paths ................. None........................default
  test_data_weights ............... None........................default
  tokenizer_type .................. GPT2BPETokenizer............default
  top_k ........................... 0...........................default
  top_p ........................... 0.0.........................default
  train_data_paths ................ None........................default
  train_data_weights .............. None........................default
  use_bnb_optimizer ............... False.......................default
  use_checkpoint_lr_scheduler ..... False.......................default
  use_cpu_initialization .......... False.......................default
  valid_data_paths ................ None........................default
  valid_data_weights .............. None........................default
  wandb_host ...................... https://api.wandb.ai........default
  wandb_project ................... neox........................default
  wandb_team ...................... None........................default
  warmup .......................... 0.01........................default
  weight_by_num_documents ......... False.......................default
  weighted_sampler_alpha .......... 0.3.........................default
  world_size ...................... None........................default
---------------- end of arguments ----------------
[2022-07-21 05:12:58,859] [WARNING] [runner.py:126:fetch_hostfile] Unable to find hostfile, will proceed with training with local resources only.
[2022-07-21 05:12:58,860] [INFO] [runner.py:366:main] cmd = /usr/bin/python -u -m deepspeed.launcher.launch --world_info=eyJsb2NhbGhvc3QiOiBbMCwgMV19 --master_addr=127.0.0.1 --master_port=29500 generate.py --deepspeed_config {"train_batch_size": 64, "train_micro_batch_size_per_gpu": 8, "gradient_accumulation_steps": 4, "optimizer": {"type": "adam", "params": {"lr": 0.00016, "betas": [0.9, 0.999], "eps": 1e-08}}, "fp16": {"fp16": true, "enabled": true, "loss_scale": 0, "initial_scale_power": 16, "loss_scale_window": 1000, "hysteresis": 2, "min_loss_scale": 1}, "gradient_clipping": 1.0, "zero_optimization": {"stage": 1, "allgather_partitions": true, "allgather_bucket_size": 500000000, "overlap_comm": true, "reduce_scatter": true, "reduce_bucket_size": 500000000, "contiguous_gradients": true, "cpu_offload": false}, "wall_clock_breakdown": true, "zero_allow_untested_optimizer": true} --megatron_config {"train_batch_size": 64, "train_micro_batch_size_per_gpu": 8, "gradient_accumulation_steps": 4, "optimizer": {"type": "adam", "params": {"lr": 0.00016, "betas": [0.9, 0.999], "eps": 1e-08}}, "fp16": {"fp16": true, "enabled": true, "loss_scale": 0, "initial_scale_power": 16, "loss_scale_window": 1000, "hysteresis": 2, "min_loss_scale": 1}, "gradient_clipping": 1.0, "zero_optimization": {"stage": 1, "allgather_partitions": true, "allgather_bucket_size": 500000000, "overlap_comm": true, "reduce_scatter": true, "reduce_bucket_size": 500000000, "contiguous_gradients": true, "cpu_offload": false}, "wall_clock_breakdown": true, "zero_allow_untested_optimizer": true, "precision": "fp16", "num_layers": 32, "hidden_size": 2560, "num_attention_heads": 32, "seq_length": 2048, "max_position_embeddings": 2048, "pos_emb": "rotary", "no_weight_tying": true, "attention_config": ["global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global"], "sparsity_config": {}, "scaled_upper_triang_masked_softmax_fusion": true, "bias_gelu_fusion": true, "lr_decay_style": "cosine", "lr_decay_iters": 160000, "zero_stage": 1, "zero_reduce_scatter": true, "zero_contiguous_gradients": true, "zero_reduce_bucket_size": 500000000, "zero_allgather_bucket_size": 500000000, "lr": 0.00016, "data_path": "data/code/code_text_document", "data_impl": "mmap", "save": "checkpoints", "config_files": {"text_generation.yml": "# Parameters used for text generation\n# Make sure `load` is specified somewhere else\n{\n  # Text gen type: `input-file`, `unconditional` or `interactive`\n  \"text-gen-type\": \"interactive\",\n \n  # Params for all\n  \"maximum_tokens\": 256,\n  \"temperature\": 0.5,\n  \"top_p\": 0.0,\n  \"top_k\": 0,\n  \"recompute\": false,\n  \n  # `unconditional`: samples\n  \"num-samples\": 10,\n\n  # input/output file\n  \"sample-input-file\": \"sample_input.txt\",\n  \"sample-output-file\": \"sample_output.txt\",\n}", "local_setup.yml": "# Suggested data paths when using GPT-NeoX locally\n{\n  \"data-path\": \"data/code/code_text_document\",\n  \n  # or for weighted datasets: \n  # \"train-data-paths\": [\"data/enron/enron_text_document\", \"data/enron/enron_text_document\"],\n  # \"test-data-paths\": [\"data/enron/enron_text_document\", \"data/enron/enron_text_document\"],\n  # \"valid-data-paths\": [\"data/enron/enron_text_document\", \"data/enron/enron_text_document\"],\n  # \"train-data-weights\": [1., 2.],\n  # \"test-data-weights\": [2., 1.],\n  # \"valid-data-weights\": [0.5, 0.4],\n\n  # If weight_by_num_documents is True, Builds dataset weights from a multinomial distribution over groups of data according to the number of documents in each group. \n  # WARNING: setting this to True will override any user provided weights\n  # \"weight_by_num_documents\": false,\n  # \"weighted_sampler_alpha\": 0.3,\n\n  \"vocab-file\": \"data/code-vocab.json\",\n  \"merge-file\": \"data/code-merges.txt\",\n\n  \"save\": \"checkpoints\",\n  \"load\": \"checkpoints\",\n  \"checkpoint_validation_with_forward_pass\": False,\n  \n  \"tensorboard-dir\": \"tensorboard\",\n  \"log-dir\": \"logs\",\n  \"use_wandb\": True,\n  \"wandb_host\": \"https://api.wandb.ai\",\n  \"wandb_project\": \"neox\"\n}", "2-7B.yml": "# GPT-2 pretraining setup\n{\n   # parallelism settings ( you will want to change these based on your cluster setup, ideally scheduling pipeline stages\n   # across the node boundaries )\n   \"pipe-parallel-size\": 1,\n   \"model-parallel-size\": 1,\n\n   # model settings\n   \"num-layers\": 32,\n   \"hidden-size\": 2560,\n   \"num-attention-heads\": 32,\n   \"seq-length\": 2048,\n   \"max-position-embeddings\": 2048,\n   \"norm\": \"layernorm\",\n   \"pos-emb\": \"rotary\",\n   \"no-weight-tying\": true,\n\n   # these should provide some speedup but takes a while to build, set to true if desired\n   \"scaled-upper-triang-masked-softmax-fusion\": true,\n   \"bias-gelu-fusion\": true,\n\n   # optimizer settings\n   \"zero_allow_untested_optimizer\": true,\n   \"optimizer\": {\n     \"type\": \"adam\",\n     \"params\": {\n       \"lr\": 0.00016,\n       \"betas\": [0.9, 0.999],\n       \"eps\": 1.0e-8,\n     }\n   },\n   \"zero_optimization\": {\n    \"stage\": 1,\n    \"allgather_partitions\": True,\n    \"allgather_bucket_size\": 500000000,\n    \"overlap_comm\": True,\n    \"reduce_scatter\": True,\n    \"reduce_bucket_size\": 500000000,\n    \"contiguous_gradients\": True,\n    \"cpu_offload\": False\n  },\n\n   # batch / data settings\n   \"train_micro_batch_size_per_gpu\": 8,\n   \"gradient_accumulation_steps\": 4,\n   \"data-impl\": \"mmap\",\n   \"split\": \"989,10,1\",\n\n   # activation checkpointing\n   \"checkpoint-activations\": true,\n   \"checkpoint-num-layers\": 1,\n   \"partition-activations\": true,\n   \"synchronize-each-layer\": true,\n\n   # regularization\n   \"gradient_clipping\": 1.0,\n   \"weight-decay\": 0,\n   \"hidden-dropout\": 0,\n   \"attention-dropout\": 0,\n\n   # precision settings\n   \"fp16\": { \n     \"fp16\": true,\n     \"enabled\": true,\n     \"loss_scale\": 0,\n     \"initial_scale_power\": 16,\n     \"loss_scale_window\": 1000,\n     \"hysteresis\": 2,\n     \"min_loss_scale\": 1\n   },\n\n   # misc. training settings\n   \"train-iters\": 160000,\n   \"lr-decay-iters\": 160000,\n   \"distributed-backend\": \"nccl\",\n   \"lr-decay-style\": \"cosine\",\n   \"warmup\": 0.01,\n   \"save-interval\": 1000,\n   \"eval-interval\": 1000,\n   \"eval-iters\": 10,\n\n   # logging\n   \"log-interval\": 100,\n   \"steps_per_print\": 10,\n   \"keep-last-n-checkpoints\": 1,\n   \"wall_clock_breakdown\": true,\n}\n"}, "load": "checkpoints", "save_interval": 1000, "batch_size": 8, "train_iters": 160000, "eval_iters": 10, "keep_last_n_checkpoints": 1, "split": "989,10,1", "vocab_file": "data/code-vocab.json", "merge_file": "data/code-merges.txt", "attention_dropout": 0, "hidden_dropout": 0, "weight_decay": 0, "checkpoint_activations": true, "synchronize_each_layer": true, "partition_activations": true, "gas": 4, "clip_grad": 1.0, "dynamic_loss_scale": true, "pipe_parallel_size": 1, "is_pipe_parallel": true, "use_wandb": true, "wandb_group": "jtRPtjruy7PQkWHayfg7cH_6sweym4s", "log_dir": "logs", "tensorboard_dir": "tensorboard", "log_interval": 100, "text_gen_type": "interactive", "temperature": 0.5, "maximum_tokens": 256, "sample_input_file": "sample_input.txt", "sample_output_file": "sample_output.txt", "num_samples": 10, "user_script": "generate.py", "global_num_gpus": 2}
[2022-07-21 05:12:59,743] [INFO] [launch.py:82:main] WORLD INFO DICT: {'localhost': [0, 1]}
[2022-07-21 05:12:59,743] [INFO] [launch.py:88:main] nnodes=1, num_local_procs=2, node_rank=0
[2022-07-21 05:12:59,743] [INFO] [launch.py:103:main] global_rank_mapping=defaultdict(<class 'list'>, {'localhost': [0, 1]})
[2022-07-21 05:12:59,743] [INFO] [launch.py:104:main] dist_world_size=2
[2022-07-21 05:12:59,743] [INFO] [launch.py:112:main] Setting CUDA_VISIBLE_DEVICES=0,1
NeoXArgs.configure_distributed_args() using world size: 2 and model-parallel size: 1 
> building GPT2BPETokenizer tokenizer ...
 > padded vocab (size: 50257) with 47 dummy tokens (new size: 50304)
> initializing torch distributed ...
[2022-07-21 05:13:02,390] [INFO] [distributed.py:46:init_distributed] Initializing torch distributed with backend: nccl
[2022-07-21 05:13:02,482] [INFO] [distributed.py:46:init_distributed] Initializing torch distributed with backend: nccl
> initializing model parallel with size 1
MPU DP: [0, 1]
MPU PP: [0]
MPU PP: [1]
MPU MP: [0]
MPU MP: [1]
> setting random seeds to 1234 ...
[2022-07-21 05:13:02,518] [INFO] [checkpointing.py:223:model_parallel_cuda_manual_seed] > initializing model parallel cuda seeds on global rank 0, model parallel rank 0, and data parallel rank 0 with model parallel seed: 3952 and data parallel seed: 1234
make: Entering directory '/gpt-neox/megatron/data'
make: Nothing to be done for 'default'.
make: Leaving directory '/gpt-neox/megatron/data'
building GPT2 model ...
SEED_LAYERS=False BASE_SEED=1234 SEED_FN=None
Using topology: {ProcessCoord(pipe=0, data=0, model=0): 0, ProcessCoord(pipe=0, data=1, model=0): 1}
[2022-07-21 05:13:02,651] [INFO] [module.py:363:_partition_layers] Partitioning pipeline stages with method type:transformer|mlp
stage=0 layers=37
     0: EmbeddingPipe
     1: _pre_transformer_block
     2: ParallelTransformerLayerPipe
     3: ParallelTransformerLayerPipe
     4: ParallelTransformerLayerPipe
     5: ParallelTransformerLayerPipe
     6: ParallelTransformerLayerPipe
     7: ParallelTransformerLayerPipe
     8: ParallelTransformerLayerPipe
     9: ParallelTransformerLayerPipe
    10: ParallelTransformerLayerPipe
    11: ParallelTransformerLayerPipe
    12: ParallelTransformerLayerPipe
    13: ParallelTransformerLayerPipe
    14: ParallelTransformerLayerPipe
    15: ParallelTransformerLayerPipe
    16: ParallelTransformerLayerPipe
    17: ParallelTransformerLayerPipe
    18: ParallelTransformerLayerPipe
    19: ParallelTransformerLayerPipe
    20: ParallelTransformerLayerPipe
    21: ParallelTransformerLayerPipe
    22: ParallelTransformerLayerPipe
    23: ParallelTransformerLayerPipe
    24: ParallelTransformerLayerPipe
    25: ParallelTransformerLayerPipe
    26: ParallelTransformerLayerPipe
    27: ParallelTransformerLayerPipe
    28: ParallelTransformerLayerPipe
    29: ParallelTransformerLayerPipe
    30: ParallelTransformerLayerPipe
    31: ParallelTransformerLayerPipe
    32: ParallelTransformerLayerPipe
    33: ParallelTransformerLayerPipe
    34: _post_transformer_block
    35: NormPipe
    36: ParallelLinearPipe
DeepSpeed is enabled.
[2022-07-21 05:13:05,069] [INFO] [logging.py:60:log_dist] [Rank 0] DeepSpeed info: version=0.3.15+eb7f5cf, git-hash=eb7f5cf, git-branch=main
[2022-07-21 05:13:05,070] [WARNING] [config.py:77:_sanity_check] DeepSpeedConfig: cpu_offload is deprecated. Please use offload_optimizer.
[2022-07-21 05:13:05,102] [WARNING] [config.py:77:_sanity_check] DeepSpeedConfig: cpu_offload is deprecated. Please use offload_optimizer.
[2022-07-21 05:13:05,172] [INFO] [config.py:759:print] DeepSpeedEngine configuration:
[2022-07-21 05:13:05,173] [INFO] [config.py:763:print]   activation_checkpointing_config  {
    "partition_activations": false, 
    "contiguous_memory_optimization": false, 
    "cpu_checkpointing": false, 
    "number_checkpoints": null, 
    "synchronize_checkpoint_boundary": false, 
    "profile": false
}
[2022-07-21 05:13:05,173] [INFO] [config.py:763:print]   aio_config ................... {'block_size': 1048576, 'queue_depth': 8, 'thread_count': 1, 'single_submit': False, 'overlap_events': True}
[2022-07-21 05:13:05,173] [INFO] [config.py:763:print]   allreduce_always_fp32 ........ False
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   amp_enabled .................. False
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   amp_params ................... False
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   checkpoint_tag_validation_enabled  True
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   checkpoint_tag_validation_fail  False
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   disable_allgather ............ False
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   dump_state ................... False
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   dynamic_loss_scale_args ...... {'init_scale': 65536, 'scale_window': 1000, 'delayed_shift': 2, 'min_scale': 1}
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   elasticity_enabled ........... False
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   flops_profiler_config ........ {
    "enabled": false, 
    "profile_step": 1, 
    "module_depth": -1, 
    "top_modules": 3, 
    "detailed": true
}
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   fp16_enabled ................. True
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   fp16_type .................... fp16
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   global_rank .................. 0
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   gradient_accumulation_steps .. 4
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   gradient_clipping ............ 1.0
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   gradient_predivide_factor .... 1.0
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   initial_dynamic_scale ........ 65536
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   loss_scale ................... 0
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   memory_breakdown ............. False
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   optimizer_legacy_fusion ...... False
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   optimizer_name ............... adam
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   optimizer_params ............. {'lr': 0.00016, 'betas': [0.9, 0.999], 'eps': 1e-08}
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   pipeline ..................... {'stages': 'auto', 'partition': 'best', 'seed_layers': False, 'activation_checkpoint_interval': 0}
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   pld_enabled .................. False
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   pld_params ................... False
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   precision .................... torch.float16
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   prescale_gradients ........... False
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   scheduler_name ............... None
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   scheduler_params ............. None
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   sparse_attention ............. None
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   sparse_gradients_enabled ..... False
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   steps_per_print .............. 10
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   tensorboard_enabled .......... False
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   tensorboard_job_name ......... DeepSpeedJobName
[2022-07-21 05:13:05,174] [INFO] [config.py:763:print]   tensorboard_output_path ...... 
[2022-07-21 05:13:05,175] [INFO] [config.py:763:print]   train_batch_size ............. 64
[2022-07-21 05:13:05,175] [INFO] [config.py:763:print]   train_micro_batch_size_per_gpu  8
[2022-07-21 05:13:05,175] [INFO] [config.py:763:print]   wall_clock_breakdown ......... True
[2022-07-21 05:13:05,175] [INFO] [config.py:763:print]   world_size ................... 2
[2022-07-21 05:13:05,175] [INFO] [config.py:763:print]   zero_allow_untested_optimizer  True
[2022-07-21 05:13:05,175] [INFO] [config.py:763:print]   zero_config .................. {
    "stage": 0, 
    "contiguous_gradients": false, 
    "reduce_scatter": true, 
    "reduce_bucket_size": 5.000000e+08, 
    "allgather_partitions": true, 
    "allgather_bucket_size": 5.000000e+08, 
    "overlap_comm": false, 
    "load_from_fp32_weights": true, 
    "elastic_checkpoint": true, 
    "offload_param": null, 
    "offload_optimizer": null, 
    "sub_group_size": 1.000000e+12, 
    "prefetch_bucket_size": 5.000000e+07, 
    "param_persistence_threshold": 1.000000e+05, 
    "max_live_parameters": 1.000000e+09, 
    "max_reuse_distance": 1.000000e+09, 
    "gather_fp16_weights_on_model_save": false
}
[2022-07-21 05:13:05,175] [INFO] [config.py:763:print]   zero_enabled ................. False
[2022-07-21 05:13:05,175] [INFO] [config.py:763:print]   zero_optimization_stage ...... 0
[2022-07-21 05:13:05,175] [INFO] [config.py:765:print]   json = {
    "train_batch_size": 64, 
    "train_micro_batch_size_per_gpu": 8, 
    "gradient_accumulation_steps": 4, 
    "optimizer": {
        "type": "adam", 
        "params": {
            "lr": 0.00016, 
            "betas": [0.9, 0.999], 
            "eps": 1e-08
        }
    }, 
    "fp16": {
        "fp16": true, 
        "enabled": true, 
        "loss_scale": 0, 
        "initial_scale_power": 16, 
        "loss_scale_window": 1000, 
        "hysteresis": 2, 
        "min_loss_scale": 1
    }, 
    "gradient_clipping": 1.0, 
    "zero_optimization": {
        "stage": 0, 
        "allgather_partitions": true, 
        "reduce_scatter": true, 
        "allgather_bucket_size": 5.000000e+08, 
        "overlap_comm": false, 
        "reduce_bucket_size": 5.000000e+08, 
        "contiguous_gradients": false, 
        "cpu_offload": false
    }, 
    "wall_clock_breakdown": true, 
    "zero_allow_untested_optimizer": true
}
Using /root/.cache/torch_extensions as PyTorch extensions root...
Using /root/.cache/torch_extensions as PyTorch extensions root...
Emitting ninja build file /root/.cache/torch_extensions/utils/build.ninja...
Building extension module utils...
Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
ninja: no work to do.
Loading extension module utils...
Time to load utils op: 0.37787580490112305 seconds
[2022-07-21 05:13:05,558] [INFO] [engine.py:84:__init__] CONFIG: micro_batches=4 micro_batch_size=8
Loading extension module utils...
Time to load utils op: 0.40610766410827637 seconds
[2022-07-21 05:13:05,679] [INFO] [engine.py:141:__init__] RANK=0 STAGE=0 LAYERS=37 [0, 37) STAGE_PARAMS=2775208960 (2775.209M) TOTAL_PARAMS=2775208960 (2775.209M) UNIQUE_PARAMS=2775208960 (2775.209M)
 > number of parameters on model parallel rank 0: 2775208960
 > total params: 2,775,208,960
[2022-07-21 05:13:05,702] [INFO] [engine.py:1551:_load_checkpoint] rank: 0 loading checkpoint: checkpoints/global_step150000/mp_rank_00_model_states.pt
[2022-07-21 05:13:05,702] [INFO] [engine.py:1551:_load_checkpoint] rank: 1 loading checkpoint: checkpoints/global_step150000/mp_rank_00_model_states.pt
[2022-07-21 05:13:05,901] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=0 file=checkpoints/global_step150000/layer_00-model_00-model_states.pt
[2022-07-21 05:13:06,022] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=2 file=checkpoints/global_step150000/layer_02-model_00-model_states.pt
[2022-07-21 05:13:06,138] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=3 file=checkpoints/global_step150000/layer_03-model_00-model_states.pt
[2022-07-21 05:13:06,254] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=4 file=checkpoints/global_step150000/layer_04-model_00-model_states.pt
[2022-07-21 05:13:06,370] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=5 file=checkpoints/global_step150000/layer_05-model_00-model_states.pt
[2022-07-21 05:13:06,481] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=6 file=checkpoints/global_step150000/layer_06-model_00-model_states.pt
[2022-07-21 05:13:06,592] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=7 file=checkpoints/global_step150000/layer_07-model_00-model_states.pt
[2022-07-21 05:13:06,730] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=8 file=checkpoints/global_step150000/layer_08-model_00-model_states.pt
[2022-07-21 05:13:06,854] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=9 file=checkpoints/global_step150000/layer_09-model_00-model_states.pt
[2022-07-21 05:13:06,968] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=10 file=checkpoints/global_step150000/layer_10-model_00-model_states.pt
[2022-07-21 05:13:07,083] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=11 file=checkpoints/global_step150000/layer_11-model_00-model_states.pt
[2022-07-21 05:13:07,199] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=12 file=checkpoints/global_step150000/layer_12-model_00-model_states.pt
[2022-07-21 05:13:07,313] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=13 file=checkpoints/global_step150000/layer_13-model_00-model_states.pt
[2022-07-21 05:13:07,433] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=14 file=checkpoints/global_step150000/layer_14-model_00-model_states.pt
[2022-07-21 05:13:07,550] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=15 file=checkpoints/global_step150000/layer_15-model_00-model_states.pt
[2022-07-21 05:13:07,667] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=16 file=checkpoints/global_step150000/layer_16-model_00-model_states.pt
[2022-07-21 05:13:07,782] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=17 file=checkpoints/global_step150000/layer_17-model_00-model_states.pt
[2022-07-21 05:13:07,899] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=18 file=checkpoints/global_step150000/layer_18-model_00-model_states.pt
[2022-07-21 05:13:08,007] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=19 file=checkpoints/global_step150000/layer_19-model_00-model_states.pt
[2022-07-21 05:13:08,142] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=20 file=checkpoints/global_step150000/layer_20-model_00-model_states.pt
[2022-07-21 05:13:08,251] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=21 file=checkpoints/global_step150000/layer_21-model_00-model_states.pt
[2022-07-21 05:13:08,358] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=22 file=checkpoints/global_step150000/layer_22-model_00-model_states.pt
[2022-07-21 05:13:08,466] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=23 file=checkpoints/global_step150000/layer_23-model_00-model_states.pt
[2022-07-21 05:13:08,574] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=24 file=checkpoints/global_step150000/layer_24-model_00-model_states.pt
[2022-07-21 05:13:08,681] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=25 file=checkpoints/global_step150000/layer_25-model_00-model_states.pt
[2022-07-21 05:13:08,786] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=26 file=checkpoints/global_step150000/layer_26-model_00-model_states.pt
[2022-07-21 05:13:08,894] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=27 file=checkpoints/global_step150000/layer_27-model_00-model_states.pt
[2022-07-21 05:13:09,003] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=28 file=checkpoints/global_step150000/layer_28-model_00-model_states.pt
[2022-07-21 05:13:09,114] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=29 file=checkpoints/global_step150000/layer_29-model_00-model_states.pt
[2022-07-21 05:13:09,222] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=30 file=checkpoints/global_step150000/layer_30-model_00-model_states.pt
[2022-07-21 05:13:09,332] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=31 file=checkpoints/global_step150000/layer_31-model_00-model_states.pt
[2022-07-21 05:13:09,438] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=32 file=checkpoints/global_step150000/layer_32-model_00-model_states.pt
[2022-07-21 05:13:09,544] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=33 file=checkpoints/global_step150000/layer_33-model_00-model_states.pt
[2022-07-21 05:13:09,544] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=35 file=checkpoints/global_step150000/layer_35-model_00-model_states.pt
[2022-07-21 05:13:09,752] [INFO] [module.py:576:load_state_dir] RANK=0 Loaded layer=36 file=checkpoints/global_step150000/layer_36-model_00-model_states.pt
 > validated currently set args with arguments in the checkpoint ...
  successfully loaded checkpoints/global_step150000/mp_rank_00_model_states.pt
Loading checkpoint and starting from iteration 150000
Finished loading model
[H[2J[3JContext prompt >>> def return1():\n """Returns 1."""\n

Traceback (most recent call last):
  File "generate.py", line 74, in <module>
    main()
  File "generate.py", line 59, in main
    generate_samples_interactive(
  File "/gpt-neox/megatron/text_generation_utils.py", line 751, in generate_samples_interactive
    for (
  File "/gpt-neox/megatron/text_generation_utils.py", line 317, in stream_tokens
    logits, layer_past = forward_model(neox_args, model, model_inputs)
  File "/gpt-neox/megatron/text_generation_utils.py", line 137, in forward_model
    return model.module(model_inputs)
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/usr/local/lib/python3.8/dist-packages/deepspeed/runtime/pipe/module.py", line 335, in forward
    x = func(forward_input)
  File "/usr/local/lib/python3.8/dist-packages/deepspeed/runtime/pipe/module.py", line 328, in exec_func
    inputs = layer(inputs)
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/gpt-neox/megatron/model/transformer.py", line 686, in forward
    outputs = super().forward(hidden_states, attention_mask, layer_past=past)
  File "/gpt-neox/megatron/model/transformer.py", line 639, in forward
    attention_output, attention_bias = self.attention(
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/gpt-neox/megatron/model/transformer.py", line 516, in forward
    output, bias = self.dense(context_layer)
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/gpt-neox/megatron/mpu/layers.py", line 446, in forward
    output_parallel = F.linear(input_parallel, self.weight)
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/functional.py", line 1753, in linear
    return torch._C._nn.linear(input, weight, bias)
RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling `cublasGemmEx( handle, opa, opb, m, n, k, &falpha, a, CUDA_R_16F, lda, b, CUDA_R_16F, ldb, &fbeta, c, CUDA_R_16F, ldc, CUDA_R_32F, CUBLAS_GEMM_DFALT_TENSOR_OP)`
Traceback (most recent call last):
  File "generate.py", line 74, in <module>
    main()
  File "generate.py", line 59, in main
    generate_samples_interactive(
  File "/gpt-neox/megatron/text_generation_utils.py", line 751, in generate_samples_interactive
    for (
  File "/gpt-neox/megatron/text_generation_utils.py", line 317, in stream_tokens
    logits, layer_past = forward_model(neox_args, model, model_inputs)
  File "/gpt-neox/megatron/text_generation_utils.py", line 137, in forward_model
    return model.module(model_inputs)
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/usr/local/lib/python3.8/dist-packages/deepspeed/runtime/pipe/module.py", line 335, in forward
    x = func(forward_input)
  File "/usr/local/lib/python3.8/dist-packages/deepspeed/runtime/pipe/module.py", line 328, in exec_func
    inputs = layer(inputs)
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/gpt-neox/megatron/model/transformer.py", line 686, in forward
    outputs = super().forward(hidden_states, attention_mask, layer_past=past)
  File "/gpt-neox/megatron/model/transformer.py", line 639, in forward
    attention_output, attention_bias = self.attention(
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/gpt-neox/megatron/model/transformer.py", line 516, in forward
    output, bias = self.dense(context_layer)
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/gpt-neox/megatron/mpu/layers.py", line 446, in forward
    output_parallel = F.linear(input_parallel, self.weight)
  File "/usr/local/lib/python3.8/dist-packages/torch/nn/functional.py", line 1753, in linear
    return torch._C._nn.linear(input, weight, bias)
RuntimeError: CUDA error: CUBLAS_STATUS_EXECUTION_FAILED when calling `cublasGemmEx( handle, opa, opb, m, n, k, &falpha, a, CUDA_R_16F, lda, b, CUDA_R_16F, ldb, &fbeta, c, CUDA_R_16F, ldc, CUDA_R_32F, CUBLAS_GEMM_DFALT_TENSOR_OP)`
Killing subprocess 1054
Killing subprocess 1055
Traceback (most recent call last):
  File "/usr/lib/python3.8/runpy.py", line 194, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "/usr/lib/python3.8/runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "/usr/local/lib/python3.8/dist-packages/deepspeed/launcher/launch.py", line 179, in <module>
    main()
  File "/usr/local/lib/python3.8/dist-packages/deepspeed/launcher/launch.py", line 169, in main
    sigkill_handler(signal.SIGTERM, None)  # not coming back
  File "/usr/local/lib/python3.8/dist-packages/deepspeed/launcher/launch.py", line 147, in sigkill_handler
    raise subprocess.CalledProcessError(returncode=last_return_code, cmd=cmd)
subprocess.CalledProcessError: Command '['/usr/bin/python', '-u', 'generate.py', '--local_rank=1', '--deepspeed_config', '{"train_batch_size": 64, "train_micro_batch_size_per_gpu": 8, "gradient_accumulation_steps": 4, "optimizer": {"type": "adam", "params": {"lr": 0.00016, "betas": [0.9, 0.999], "eps": 1e-08}}, "fp16": {"fp16": true, "enabled": true, "loss_scale": 0, "initial_scale_power": 16, "loss_scale_window": 1000, "hysteresis": 2, "min_loss_scale": 1}, "gradient_clipping": 1.0, "zero_optimization": {"stage": 1, "allgather_partitions": true, "allgather_bucket_size": 500000000, "overlap_comm": true, "reduce_scatter": true, "reduce_bucket_size": 500000000, "contiguous_gradients": true, "cpu_offload": false}, "wall_clock_breakdown": true, "zero_allow_untested_optimizer": true}', '--megatron_config', '{"train_batch_size": 64, "train_micro_batch_size_per_gpu": 8, "gradient_accumulation_steps": 4, "optimizer": {"type": "adam", "params": {"lr": 0.00016, "betas": [0.9, 0.999], "eps": 1e-08}}, "fp16": {"fp16": true, "enabled": true, "loss_scale": 0, "initial_scale_power": 16, "loss_scale_window": 1000, "hysteresis": 2, "min_loss_scale": 1}, "gradient_clipping": 1.0, "zero_optimization": {"stage": 1, "allgather_partitions": true, "allgather_bucket_size": 500000000, "overlap_comm": true, "reduce_scatter": true, "reduce_bucket_size": 500000000, "contiguous_gradients": true, "cpu_offload": false}, "wall_clock_breakdown": true, "zero_allow_untested_optimizer": true, "precision": "fp16", "num_layers": 32, "hidden_size": 2560, "num_attention_heads": 32, "seq_length": 2048, "max_position_embeddings": 2048, "pos_emb": "rotary", "no_weight_tying": true, "attention_config": ["global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global"], "sparsity_config": {}, "scaled_upper_triang_masked_softmax_fusion": true, "bias_gelu_fusion": true, "lr_decay_style": "cosine", "lr_decay_iters": 160000, "zero_stage": 1, "zero_reduce_scatter": true, "zero_contiguous_gradients": true, "zero_reduce_bucket_size": 500000000, "zero_allgather_bucket_size": 500000000, "lr": 0.00016, "data_path": "data/code/code_text_document", "data_impl": "mmap", "save": "checkpoints", "config_files": {"text_generation.yml": "# Parameters used for text generation\\n# Make sure `load` is specified somewhere else\\n{\\n  # Text gen type: `input-file`, `unconditional` or `interactive`\\n  \\"text-gen-type\\": \\"interactive\\",\\n \\n  # Params for all\\n  \\"maximum_tokens\\": 256,\\n  \\"temperature\\": 0.5,\\n  \\"top_p\\": 0.0,\\n  \\"top_k\\": 0,\\n  \\"recompute\\": false,\\n  \\n  # `unconditional`: samples\\n  \\"num-samples\\": 10,\\n\\n  # input/output file\\n  \\"sample-input-file\\": \\"sample_input.txt\\",\\n  \\"sample-output-file\\": \\"sample_output.txt\\",\\n}", "local_setup.yml": "# Suggested data paths when using GPT-NeoX locally\\n{\\n  \\"data-path\\": \\"data/code/code_text_document\\",\\n  \\n  # or for weighted datasets: \\n  # \\"train-data-paths\\": [\\"data/enron/enron_text_document\\", \\"data/enron/enron_text_document\\"],\\n  # \\"test-data-paths\\": [\\"data/enron/enron_text_document\\", \\"data/enron/enron_text_document\\"],\\n  # \\"valid-data-paths\\": [\\"data/enron/enron_text_document\\", \\"data/enron/enron_text_document\\"],\\n  # \\"train-data-weights\\": [1., 2.],\\n  # \\"test-data-weights\\": [2., 1.],\\n  # \\"valid-data-weights\\": [0.5, 0.4],\\n\\n  # If weight_by_num_documents is True, Builds dataset weights from a multinomial distribution over groups of data according to the number of documents in each group. \\n  # WARNING: setting this to True will override any user provided weights\\n  # \\"weight_by_num_documents\\": false,\\n  # \\"weighted_sampler_alpha\\": 0.3,\\n\\n  \\"vocab-file\\": \\"data/code-vocab.json\\",\\n  \\"merge-file\\": \\"data/code-merges.txt\\",\\n\\n  \\"save\\": \\"checkpoints\\",\\n  \\"load\\": \\"checkpoints\\",\\n  \\"checkpoint_validation_with_forward_pass\\": False,\\n  \\n  \\"tensorboard-dir\\": \\"tensorboard\\",\\n  \\"log-dir\\": \\"logs\\",\\n  \\"use_wandb\\": True,\\n  \\"wandb_host\\": \\"https://api.wandb.ai\\",\\n  \\"wandb_project\\": \\"neox\\"\\n}", "2-7B.yml": "# GPT-2 pretraining setup\\n{\\n   # parallelism settings ( you will want to change these based on your cluster setup, ideally scheduling pipeline stages\\n   # across the node boundaries )\\n   \\"pipe-parallel-size\\": 1,\\n   \\"model-parallel-size\\": 1,\\n\\n   # model settings\\n   \\"num-layers\\": 32,\\n   \\"hidden-size\\": 2560,\\n   \\"num-attention-heads\\": 32,\\n   \\"seq-length\\": 2048,\\n   \\"max-position-embeddings\\": 2048,\\n   \\"norm\\": \\"layernorm\\",\\n   \\"pos-emb\\": \\"rotary\\",\\n   \\"no-weight-tying\\": true,\\n\\n   # these should provide some speedup but takes a while to build, set to true if desired\\n   \\"scaled-upper-triang-masked-softmax-fusion\\": true,\\n   \\"bias-gelu-fusion\\": true,\\n\\n   # optimizer settings\\n   \\"zero_allow_untested_optimizer\\": true,\\n   \\"optimizer\\": {\\n     \\"type\\": \\"adam\\",\\n     \\"params\\": {\\n       \\"lr\\": 0.00016,\\n       \\"betas\\": [0.9, 0.999],\\n       \\"eps\\": 1.0e-8,\\n     }\\n   },\\n   \\"zero_optimization\\": {\\n    \\"stage\\": 1,\\n    \\"allgather_partitions\\": True,\\n    \\"allgather_bucket_size\\": 500000000,\\n    \\"overlap_comm\\": True,\\n    \\"reduce_scatter\\": True,\\n    \\"reduce_bucket_size\\": 500000000,\\n    \\"contiguous_gradients\\": True,\\n    \\"cpu_offload\\": False\\n  },\\n\\n   # batch / data settings\\n   \\"train_micro_batch_size_per_gpu\\": 8,\\n   \\"gradient_accumulation_steps\\": 4,\\n   \\"data-impl\\": \\"mmap\\",\\n   \\"split\\": \\"989,10,1\\",\\n\\n   # activation checkpointing\\n   \\"checkpoint-activations\\": true,\\n   \\"checkpoint-num-layers\\": 1,\\n   \\"partition-activations\\": true,\\n   \\"synchronize-each-layer\\": true,\\n\\n   # regularization\\n   \\"gradient_clipping\\": 1.0,\\n   \\"weight-decay\\": 0,\\n   \\"hidden-dropout\\": 0,\\n   \\"attention-dropout\\": 0,\\n\\n   # precision settings\\n   \\"fp16\\": { \\n     \\"fp16\\": true,\\n     \\"enabled\\": true,\\n     \\"loss_scale\\": 0,\\n     \\"initial_scale_power\\": 16,\\n     \\"loss_scale_window\\": 1000,\\n     \\"hysteresis\\": 2,\\n     \\"min_loss_scale\\": 1\\n   },\\n\\n   # misc. training settings\\n   \\"train-iters\\": 160000,\\n   \\"lr-decay-iters\\": 160000,\\n   \\"distributed-backend\\": \\"nccl\\",\\n   \\"lr-decay-style\\": \\"cosine\\",\\n   \\"warmup\\": 0.01,\\n   \\"save-interval\\": 1000,\\n   \\"eval-interval\\": 1000,\\n   \\"eval-iters\\": 10,\\n\\n   # logging\\n   \\"log-interval\\": 100,\\n   \\"steps_per_print\\": 10,\\n   \\"keep-last-n-checkpoints\\": 1,\\n   \\"wall_clock_breakdown\\": true,\\n}\\n"}, "load": "checkpoints", "save_interval": 1000, "batch_size": 8, "train_iters": 160000, "eval_iters": 10, "keep_last_n_checkpoints": 1, "split": "989,10,1", "vocab_file": "data/code-vocab.json", "merge_file": "data/code-merges.txt", "attention_dropout": 0, "hidden_dropout": 0, "weight_decay": 0, "checkpoint_activations": true, "synchronize_each_layer": true, "partition_activations": true, "gas": 4, "clip_grad": 1.0, "dynamic_loss_scale": true, "pipe_parallel_size": 1, "is_pipe_parallel": true, "use_wandb": true, "wandb_group": "jtRPtjruy7PQkWHayfg7cH_6sweym4s", "log_dir": "logs", "tensorboard_dir": "tensorboard", "log_interval": 100, "text_gen_type": "interactive", "temperature": 0.5, "maximum_tokens": 256, "sample_input_file": "sample_input.txt", "sample_output_file": "sample_output.txt", "num_samples": 10, "user_script": "generate.py", "global_num_gpus": 2}']' returned non-zero exit status 1.

opened by LakshyAAAgrawal 1

Context prompt - example not working

I have configured this setup in a 4GPU machine. I have done the setup using docker image. I have received the Context prompt. When I feed it an example, it is falling apart. Can you please help me to understand why it is failing?

Context prompt >>> def return1():\n """Returns 1."""\n Traceback (most recent call last): File "generate.py", line 74, in <module> main() File "generate.py", line 59, in main generate_samples_interactive( File "/gpt-neox/megatron/text_generation_utils.py", line 779, in generate_samples_interactive generated_text = neox_args.tokenizer.detokenize(generated_tokens) File "/gpt-neox/megatron/tokenizer/tokenizer.py", line 162, in detokenize return self.tokenizer.decode(token_ids) File "/gpt-neox/megatron/tokenizer/gpt2_tokenization.py", line 279, in decode text = ''.join([self.decoder[token] for token in tokens]) File "/gpt-neox/megatron/tokenizer/gpt2_tokenization.py", line 279, in <listcomp> text = ''.join([self.decoder[token] for token in tokens]) KeyError: 50269 Traceback (most recent call last): File "generate.py", line 74, in <module> main() File "generate.py", line 59, in main generate_samples_interactive( File "/gpt-neox/megatron/text_generation_utils.py", line 779, in generate_samples_interactive generated_text = neox_args.tokenizer.detokenize(generated_tokens) File "/gpt-neox/megatron/tokenizer/tokenizer.py", line 162, in detokenize return self.tokenizer.decode(token_ids) File "/gpt-neox/megatron/tokenizer/gpt2_tokenization.py", line 279, in decode text = ''.join([self.decoder[token] for token in tokens]) File "/gpt-neox/megatron/tokenizer/gpt2_tokenization.py", line 279, in <listcomp> text = ''.join([self.decoder[token] for token in tokens]) KeyError: 50269 Traceback (most recent call last): File "generate.py", line 74, in <module> main() File "generate.py", line 59, in main generate_samples_interactive( File "/gpt-neox/megatron/text_generation_utils.py", line 779, in generate_samples_interactive generated_text = neox_args.tokenizer.detokenize(generated_tokens) File "/gpt-neox/megatron/tokenizer/tokenizer.py", line 162, in detokenize return self.tokenizer.decode(token_ids) File "/gpt-neox/megatron/tokenizer/gpt2_tokenization.py", line 279, in decode text = ''.join([self.decoder[token] for token in tokens]) File "/gpt-neox/megatron/tokenizer/gpt2_tokenization.py", line 279, in <listcomp> text = ''.join([self.decoder[token] for token in tokens]) KeyError: 50269Killing subprocess 118 Killing subprocess 119Killing subprocess 120 Killing subprocess 121Traceback (most recent call last): File "/usr/lib/python3.8/runpy.py", line 194, in _run_module_as_main return _run_code(code, main_globals, None, File "/usr/lib/python3.8/runpy.py", line 87, in _run_code exec(code, run_globals) File "/usr/local/lib/python3.8/dist-packages/deepspeed/launcher/launch.py", line 179, in <module> main() File "/usr/local/lib/python3.8/dist-packages/deepspeed/launcher/launch.py", line 169, in main sigkill_handler(signal.SIGTERM, None) # not coming back File "/usr/local/lib/python3.8/dist-packages/deepspeed/launcher/launch.py", line 147, in sigkill_handler raise subprocess.CalledProcessError(returncode=last_return_code, cmd=cmd) subprocess.CalledProcessError: Command '['/usr/bin/python', '-u', 'generate.py', '--local_rank=3', '--deepspeed_config', '{"train_batch_size": 128, "train_micro_batch_size_per_gpu": 8, "gradient_accumulation_steps": 4, "optimizer": {"type": "adam", "params": {"lr": 0.00016, "betas": [0.9, 0.999], "eps": 1e-08}}, "fp16": {"fp16": true, "enabled": true, "loss_scale": 0, "initial_scale_power": 16, "loss_scale_window": 1000, "hysteresis": 2, "min_loss_scale": 1}, "gradient_clipping": 1.0, "zero_optimization": {"stage": 1, "allgather_partitions": true, "allgather_bucket_size": 500000000, "overlap_comm": true, "reduce_scatter":true, "reduce_bucket_size": 500000000, "contiguous_gradients": true, "cpu_offload": false}, "wall_clock_breakdown": true, "zero_allow_untested_optimizer": true}', '--megatron_config', '{"train_batch_size": 128, "train_micro_batch_size_per_gpu": 8, "gradient_accumulation_steps": 4, "optimizer": {"type": "adam", "params": {"lr": 0.00016, "betas": [0.9, 0.999], "eps": 1e-08}}, "fp16": {"fp16": true, "enabled": true, "loss_scale": 0, "initial_scale_power": 16, "loss_scale_window": 1000, "hysteresis": 2, "min_loss_scale": 1}, "gradient_clipping": 1.0, "zero_optimization": {"stage": 1, "allgather_partitions": true, "allgather_bucket_size": 500000000, "overlap_comm": true, "reduce_scatter": true, "reduce_bucket_size": 500000000, "contiguous_gradients": true, "cpu_offload": false}, "wall_clock_breakdown": true, "zero_allow_untested_optimizer": true, "precision": "fp16", "num_layers": 32, "hidden_size": 2560, "num_attention_heads": 32, "seq_length": 2048, "max_position_embeddings": 2048, "pos_emb": "rotary", "no_weight_tying": true, "attention_config": ["global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global", "global"], "sparsity_config": {}, "scaled_upper_triang_masked_softmax_fusion": true, "bias_gelu_fusion": true, "lr_decay_style": "cosine", "lr_decay_iters": 160000, "zero_stage": 1, "zero_reduce_scatter": true, "zero_contiguous_gradients": true, "zero_reduce_bucket_size": 500000000, "zero_allgather_bucket_size": 500000000, "lr": 0.00016, "data_path": "data/code/code_text_document", "data_impl": "mmap", "save": "checkpoints", "config_files": {"text_generation.yml": "# Parameters used for text generation\\n# Make sureloadis specified somewhere else\\n{\\n # Text gen type:input-file,unconditionalorinteractive\\n \\"text-gen-type\\": \\"interactive\\",\\n \\n # Params for all\\n \\"maximum_tokens\\": 256,\\n \\"temperature\\": 0.5,\\n \\"top_p\\": 0.0,\\n \\"top_k\\": 0,\\n \\"recompute\\": false,\\n \\n #unconditional: samples\\n \\"num-samples\\": 10,\\n\\n # input/output file\\n \\"sample-input-file\\": \\"sample_input.txt\\",\\n \\"sample-output-file\\": \\"sample_output.txt\\",\\n}", "local_setup.yml": "# Suggested data paths when using GPT-NeoX locally\\n{\\n \\"data-path\\": \\"data/code/code_text_document\\",\\n \\n # or for weighted datasets: \\n # \\"train-data-paths\\": [\\"data/enron/enron_text_document\\", \\"data/enron/enron_text_document\\"],\\n # \\"test-data-paths\\": [\\"data/enron/enron_text_document\\", \\"data/enron/enron_text_document\\"],\\n # \\"valid-data-paths\\": [\\"data/enron/enron_text_document\\", \\"data/enron/enron_text_document\\"],\\n # \\"train-data-weights\\": [1., 2.],\\n # \\"test-data-weights\\": [2., 1.],\\n # \\"valid-data-weights\\": [0.5, 0.4],\\n\\n # If weight_by_num_documents is True, Builds dataset weights from a multinomial distribution over groups of data according to the number of documents in each group. \\n # WARNING: setting this to True will override any user provided weights\\n # \\"weight_by_num_documents\\": false,\\n # \\"weighted_sampler_alpha\\": 0.3,\\n\\n \\"vocab-file\\": \\"data/code-vocab.json\\",\\n \\"merge-file\\": \\"data/code-merges.txt\\",\\n\\n \\"save\\": \\"checkpoints\\",\\n \\"load\\": \\"checkpoints\\",\\n \\"checkpoint_validation_with_forward_pass\\": False,\\n \\n \\"tensorboard-dir\\": \\"tensorboard\\",\\n \\"log-dir\\": \\"logs\\",\\n \\"use_wandb\\": True,\\n \\"wandb_host\\": \\"https://api.wandb.ai\\",\\n \\"wandb_project\\": \\"neox\\"\\n}", "2-7B.yml": "# GPT-2 pretraining setup\\n{\\n # parallelism settings ( you will want to change these based on your cluster setup, ideally scheduling pipeline stages\\n # across the node boundaries )\\n \\"pipe-parallel-size\\": 1,\\n \\"model-parallel-size\\": 1,\\n\\n # model settings\\n \\"num-layers\\": 32,\\n \\"hidden-size\\": 2560,\\n \\"num-attention-heads\\": 32,\\n \\"seq-length\\": 2048,\\n \\"max-position-embeddings\\": 2048,\\n \\"norm\\": \\"layernorm\\",\\n \\"pos-emb\\": \\"rotary\\",\\n \\"no-weight-tying\\": true,\\n\\n # these should provide some speedup but takes awhile to build, set to true if desired\\n \\"scaled-upper-triang-masked-softmax-fusion\\": true,\\n \\"bias-gelu-fusion\\": true,\\n\\n # optimizer settings\\n \\"zero_allow_untested_optimizer\\": true,\\n \\"optimizer\\": {\\n \\"type\\": \\"adam\\",\\n \\"params\\": {\\n \\"lr\\": 0.00016,\\n \\"betas\\": [0.9, 0.999],\\n \\"eps\\": 1.0e-8,\\n }\\n },\\n \\"zero_optimization\\": {\\n \\"stage\\": 1,\\n \\"allgather_partitions\\": True,\\n \\"allgather_bucket_size\\": 500000000,\\n \\"overlap_comm\\": True,\\n \\"reduce_scatter\\": True,\\n \\"reduce_bucket_size\\": 500000000,\\n \\"contiguous_gradients\\": True,\\n \\"cpu_offload\\": False\\n },\\n\\n # batch / data settings\\n \\"train_micro_batch_size_per_gpu\\": 8,\\n \\"gradient_accumulation_steps\\": 4,\\n \\"data-impl\\": \\"mmap\\",\\n \\"split\\": \\"989,10,1\\",\\n\\n # activation checkpointing\\n \\"checkpoint-activations\\": true,\\n \\"checkpoint-num-layers\\": 1,\\n \\"partition-activations\\": true,\\n \\"synchronize-each-layer\\": true,\\n\\n # regularization\\n \\"gradient_clipping\\": 1.0,\\n \\"weight-decay\\": 0,\\n \\"hidden-dropout\\": 0,\\n \\"attention-dropout\\": 0,\\n\\n # precision settings\\n \\"fp16\\": { \\n \\"fp16\\": true,\\n \\"enabled\\": true,\\n \\"loss_scale\\": 0,\\n \\"initial_scale_power\\": 16,\\n \\"loss_scale_window\\": 1000,\\n \\"hysteresis\\": 2,\\n \\"min_loss_scale\\": 1\\n },\\n\\n # misc. training settings\\n \\"train-iters\\": 160000,\\n \\"lr-decay-iters\\": 160000,\\n \\"distributed-backend\\": \\"nccl\\",\\n \\"lr-decay-style\\": \\"cosine\\",\\n \\"warmup\\": 0.01,\\n \\"save-interval\\": 1000,\\n \\"eval-interval\\": 1000,\\n \\"eval-iters\\": 10,\\n\\n # logging\\n \\"log-interval\\": 100,\\n \\"steps_per_print\\": 10,\\n \\"keep-last-n-checkpoints\\": 1,\\n \\"wall_clock_breakdown\\": true,\\n}\\n"}, "load": "checkpoints", "save_interval": 1000, "batch_size": 8, "train_iters": 160000, "eval_iters": 10, "keep_last_n_checkpoints": 1, "split": "989,10,1", "vocab_file": "data/code-vocab.json", "merge_file": "data/code-merges.txt", "attention_dropout": 0, "hidden_dropout": 0, "weight_decay": 0, "checkpoint_activations": true, "synchronize_each_layer": true, "partition_activations": true, "gas": 4, "clip_grad": 1.0, "dynamic_loss_scale": true, "pipe_parallel_size": 1, "is_pipe_parallel": true, "use_wandb": true, "wandb_group": "9j43ZWqmpkAaRAbvTjSFUt_2hatssyt", "log_dir": "logs", "tensorboard_dir": "tensorboard", "log_interval": 100, "text_gen_type": "interactive", "temperature": 0.5, "maximum_tokens": 256, "sample_input_file": "sample_input.txt", "sample_output_file": "sample_output.txt", "num_samples": 10, "user_script": "generate.py", "global_num_gpus": 4}']' returned non-zero exit status 1. mchorse@f4a108abb6e6:/gpt-neox$

opened by pallabganai 2
CUDA out of memory error on training

I was trying to train Polycoder using the preconfigured dataset, from the checkpoint checkpoints-2-7B, I used the following command as per the instructions in the repo (only changing the configs as appropriate):

sudo python ./deepy.py train.py -d configs 2-7B.yml local_setup.yml

which gave the following error:

RuntimeError: CUDA out of memory. Tried to allocate 1.86 GiB (GPU 0; 23.70 GiB total capacity; 20.49 GiB already allocated; 1.74 GiB free; 20.50 GiB reserved in total by PyTorch)

Interestingly, the full 25 Gigs of our GPU is free, as per nvidia-smi.

I tried updating the batch size, and the the only location I found to update batch size in the config files was train_micro_batch_size_per_gpu: 8, in 2-7B.yml.

It was 8, I changed it to 4, and then also to 1, but in both cases got the same error.

I am running all this in docker, as per the containerized setup instructions.

Appreciate any help!

opened by AftabHussain 3

Guide to using pre-trained large language models of source code

Related tags

Overview

Large Models of Source Code

Getting Started

Retrieving Checkpoints

From Source

Via Docker

Code Generation

Multi-lingual Models

Available Models

Training Process

Caveats

Datasets

249GB Multi-Lingual Corpus

Data Collection & Filtering

Evaluation

Evaluating Codex

Results - HumanEval

Results - Multilingual Language Modeling

Citation

Comments

Owner

Vincent Hellendoorn

RoNER is a Named Entity Recognition model based on a pre-trained BERT transformer model trained on RONECv2

Must-read papers on improving efficiency for pre-trained language models.

The repository for the paper: Multilingual Translation via Grafting Pre-trained Language Models

Prompt-learning is the latest paradigm to adapt pre-trained language models (PLMs) to downstream NLP tasks

Chinese Pre-Trained Language Models (CPM-LM) Version-I

PyTorch Implementation of "Bridging Pre-trained Language Models and Hand-crafted Features for Unsupervised POS Tagging" (Findings of ACL 2022)

Code associated with the "Data Augmentation using Pre-trained Transformer Models" paper

Silero Models: pre-trained speech-to-text, text-to-speech models and benchmarks made embarrassingly simple

Implementation of Natural Language Code Search in the project CodeBERT: A Pre-Trained Model for Programming and Natural Languages.

DziriBERT: a Pre-trained Language Model for the Algerian Dialect

Code for CodeT5: a new code-aware pre-trained encoder-decoder model.

TweebankNLP - Pre-trained Tweet NLP Pipeline (NER, tokenization, lemmatization, POS tagging, dependency parsing) + Models + Tweebank-NER

Code for text augmentation method leveraging large-scale language models

TunBERT is the first release of a pre-trained BERT model for the Tunisian dialect using a Tunisian Common-Crawl-based dataset.

One Stop Anomaly Shop: Anomaly detection using two-phase approach: (a) pre-labeling using statistics, Natural Language Processing and static rules; (b) anomaly scoring using supervised and unsupervised machine learning.

BMInf (Big Model Inference) is a low-resource inference package for large-scale pretrained language models (PLMs).

Google and Stanford University released a new pre-trained model called ELECTRA

Use PaddlePaddle to reproduce the paper：mT5: A Massively Multilingual Pre-trained Text-to-Text Transformer

ElasticBERT: A pre-trained model with multi-exit transformer architecture.