Skip to content

Latest commit

 

History

History

popxl

GPT-3 2.7B

GPT-3 2.7B for NLP pre-training and text generation, optimised for Graphcore's IPU.

Framework Domain Model Datasets Tasks Training Inference Reference
PopXL NLP GPT-3 Wikipedia Next sentence prediction, Question/Answering


Min. 64 IPUs (POD64) required


Min. 64 IPU (POD64) required

Language Models are Few-Shot Learners

Instructions summary

  1. Install and enable the Poplar SDK (see Poplar SDK setup)

  2. Install the system and Python requirements (see Environment setup)

  3. Download the WIKI-103 dataset (See Dataset setup)

Poplar SDK setup

To check if your Poplar SDK has already been enabled, run:

 echo $POPLAR_SDK_ENABLED

If no path is provided, then follow these steps:

  1. Navigate to your Poplar SDK root directory

  2. Enable the Poplar SDK with:

cd poplar-<OS version>-<SDK version>-<hash>
. enable.sh
  1. Additionally, enable PopART with:
cd popart-<OS version>-<SDK version>-<hash>
. enable.sh

More detailed instructions on setting up your Poplar environment are available in the Poplar quick start guide.

Environment setup

To prepare your environment, follow these steps:

  1. Create and activate a Python3 virtual environment:
python3 -m venv <venv name>
source <venv path>/bin/activate
  1. Navigate to this example's root directory

  2. Install the Python requirements:

pip3 install -r requirements.txt

Dataset setup

To obtain the data used for pre-training follow the below instructions.

Disk space required: 143GB - Sequence length 128 (Variable), 203GB - Sequence length 512 (Variable)

.
├── wiki_000.index
├── wiki_000.tfrecord
    .
    .
    .
├── wiki_xxx.index
└── wiki_xxx.tfrecord

0 directories, XXXX files

1. Raw data

Download the latest raw wikipedia dump using:

bash wikipedia_download.sh wikipedia_raw

Extract the data into another format:

pip3 install wikiextractor
export PYTHONIOENCODING=utf-8
export LC_ALL=C.UTF-8
bash wikipedia_extract.sh wikipedia_raw/wikidump.xml wikipedia_extracted

2. Preprocessing

Preprocess the data:

mkdir wikipedia_preprocessed
python3 wikipedia_preprocess.py --input-file-path wikipedia_extracted --output-file-path wikipedia_preprocessed

3. Generate TFRecords

To generate TFRecords from the preprocessed data

pip install tensorflow==1.15.0
mkdir wikipedia_tf
python3 write_into_tfrecord.py --input-file-path wikipedia_preprocessed/wikicorpus_en_one_article_per_line.pkl --output-file-path wikipedia_tf --seq-length 129 --stride 129

Then you need to generate the indices for the TFRecords

cd wikipedia_tf
for f in *.tfrecord; do python3 -m tfrecord.tools.tfrecord2idx $f `basename $f .tfrecord`.index; done

Custom training

Pre-training with GPT on IPU

You can run pre-training for GPT with settings defined in training.yml by using the script below. You need to provide the data files to --input_files.

python3 demo/training.py --input_files {path to your wikipedia data}/*.tfrecord

The default model size in demo pre-training is GPT-3 2.7B on POD64 (named gpt3_2.7B_pod64). You can change it to other sizes that are available in the configuration file config/training.yml using the --config CLI parameter like so.

python3 run_training.py --config gpt3_2.7B_pod64 --input_files {path to your wikipedia data}/*.tfrecord

You can run these scripts for benchmarking with generated data by executing the non-run scripts directly. For instance, the command below runs the benchmarking for GPT pre-training.

python3 training.py

When running the application, it is possible to save/load executables to/from a cache store. This allows for reusing a saved executable instead of re-compiling the model when re-running identical model configurations. To enable saving/loading from the cache store, use the environment variable POPXL_CACHE_DIR=<PATH/TO/CACHE> when running the application.

Other features

View the pre-training results in Weights & Biases

This project supports Weights & Biases, a platform to keep track of machine learning experiments. A client for Weights&Biases will be installed by default and can be used during training by passing the --wandb flag. You will need to manually log in (see the quickstart guide here) and configure the project name with --wandb-name.) For more information please see https://www.wandb.com/.

The trainings in demo are logged in wandb under project popxl-gpt. Each run has loss, learning rate and throughput logged. The version for addons and PopXL are also logged together with the configuration settings.

Configure your GPT runs

You can find configuration options for GPT in class GPTConfig in the file config/config.py. It contains configurations for these aspects:

  • Models

    You can set the parameters used in the GPT model.

    • general parameters:
      1. layers the number of decoder layers in the model,
      2. hidden_size the hidden size of the layers,
      3. sequence_length number of tokens in a sample,
      4. eval to enable the model to be built for inference or validation which will disable dropout and optimisation,
      5. dropout_prob the dropout probability,
      6. precision to set the precision used in the model parameters, for instance, popxl.float32 and popxl.float16.
      7. seed the random seed used by the model and data generation.
    • parameters for embedding layers: vocabulary size vocab_size and maximum number of positions to support in the embeddings max_positional_length.
    • parameters for attention layer: heads the number of attention heads.
  • Training

    You can configure the training options that have impact on training.

    • steps: number of steps,
    • epochs: number of epochs,
    • global_batch_size: the number of samples that contribute to an optimizer step,
    • stochastic_rounding: a flag to enable stochastic rounding,
    • optimizer: an optimizer with the following settings.
      • name: name of the optimizer, by default, AdamW.
      • learning_rate: to set up the learning rate including function used in scheduler, maximum learning rate, and warmup_proportion to set the proportion of the warmup step,
      • beta1: by default 0.9,
      • beta2: by default 0.999,
      • weight_decay: weight decay factor by default 0.0.
  • Data

    • input_files: the path to input data files
  • Execution

    It allows you to change how to execute a GPT run on IPU.

    • micro_batch_size: the number of samples that contribute to a gradient accumulation step,
    • data_parallel: the number of model replicas to use for data parallelism,
    • tensor_parallel: the number of IPUs used for tensor model parallel axis.
    • device_iterations: the number of times the training loop is executed before relinquishing control and reporting to the host,
    • io_tiles: the number of tiles dedicated to streaming data,
    • available_memory_proportion: the available memory proportion for any op that supports this option,
    • loss_scaling: the scaling factor to apply to gradients, by default 1,
    • pipeline: the pipeline layers distribution,

Note that the gradient_accumulation size is automatically computed from the global_batch_size, the micro_batch_size and data_parallel.

  • Checkpoint

    You can set the path to load and save checkpoints respectively by load and save.

Scale GPT on IPU

Here we introduce some techniques that were required to scale up the GPT model for the required capacity and throughput.

Phased Execution and RTS

For compute graphs that have memory requirements greater than the available on-chip memory, we can partition it into a series of smaller sub-graphs and execute them in series on the IPU, using remote memory to store input and output tensors between calls. This is called phased execution. We recommend the tutorial of this concept in Phased Execution in MNIST example.

In the GPT application we demonstrate this concept on a full sized model. Recomputation and replicated tensor sharding (RTS) are also used to improve the performance.

Tensor Model Parallel

Tensor-parallel training involves breaking the layers into shards, which are each allocated to a different devices. Communication is required within a layer between the different devices to rematerialise the same numerical result if tensor parallelism sharding wasn't used. For the embedding layer one all-reduces communication operations are required for the forwards and backwards pass (not included recomputation). For the GPT layers, four all-reduce operations are required for the forwards and backwards pass. For the pre-training head four all-reduce operations are required for the forwards and backwards pass.

Data Parallel

Data-parallel training involves breaking the training dataset up into multiple parts, which are each consumed by a model replica. At each optimization step, the gradients are mean-reduced across all replicas so that the weight update and model state are the same across all replicas. You can find more details about how to use data parallel in PopXL addons in MNIST example.

Pre-training code details

Constructing computational graphs for each phase

First of all, we build the training graphs for each phase, represented in the class Graphs. A phase can include one layer or consecutive layers. The execution of a phase can be for the forward graph, gradient graph, optimizer graph or a combination of them. We need to build the graphs used in each phase before we define the phases in Build the main computational graph.

The graphs required for each phase can be represented in class Graphs.

  • The fwd and bwd are respectively the forward and backward pass graphs. The bwd graph is obtained directly by using autodiff_with_accumulation from the forward graph fwd.
  • The facts has the required variable factories in the forward graph and optimizer graph. The grad_facts has the required variable factories for the backward graph.
  • The optim contains the optimizer graphs for each variable.
  • The buffers are remote buffers used to handle the loading and offloading of the activations, trainable weights, and optimiser states.
  • To handle the remote load and store for the remote buffers, we also need the:
    • graph _fwd_load that loads variables from fwd buffers and returns _fwd_load_names,
    • graph _optim_fwd_load that loads all forward and optimiser state from buffers
    • graph _optim_fwd_store that stores all forward and optimiser state to buffers
    • graph _grad_store that stores to bwd buffers. It is only used in pre-training GPT layer and task head layer.
  • To handle collectives for replica all gather and reduce replica for RTS variables, we also created the graphs:
    • graph _fwd_all_gather that does AllGather across replicas for forward RTS variables and returns _fwd_all_gather_names,
    • graph _grad_reduce that reduces across replicas for gradient RTS variables and returns _grad_reduce_names.

We created these graphs:

  • embeddings by calling the method create_embeddings_graph for the embedding layer. Note that the optimizer step for embedding layer happens straight after the backward pass on device, so there is no need to store the gradient in a buffer.
  • layer by calling the method create_decoder_block_graph for each GPT decoder layer. Its buffer contains the forward tensors and gradient tensors. Since each GPT decoder layer has identical input and output data type and shape, we stack the buffers for each layer together. Hence, the number of entries in the buffers is the same as the number of decoder layers.
  • head by calling the method create_task_head_graph for the task head layer. There are some slight differences in the implementation from the above two instances.
    • Its gradient graph is combined with the forward graph by using GPTPretrainingLossAndGrad. The calculation of gradients happens just after the forward graph calculation in the same phase. Hence, the fwd graph includes both the graph for forward pass and the calculation of its gradients.
    • Tied embedding is used. The linear layer in LM task head reuses the inputs embedding weights. As shown in the diagram below, in the forward pass the LM weights are loaded from the embedding layer weights buffer embedding.buffers.fwd.word.weight. In the backward pass, the gradient of the tied embedding weights is stored in a separate remote buffer tied_weight_grad_buffer.

Apply transformations on graphs

We then apply transformations to the graphs built:

  • recomputation: to reduce memory consumption in backward pass for embedding gradients and decoder gradients. You can transform the gradient graphs by using popxl_addons.recompute_graph method.

  • batch serialisation: to avoid the frequent loading and offloading of the variables and graphs in different layers for each batch, we use batch serialisation. It repeats the same graph with different data for each partition of the model for steps times. You can find the transformed graphs in embeddings_batch_serialise, decoder_block_batch_serialise and head_batch_serialise respectively. Each batch serialization produces the forward and gradient graphs and the activations. You can get the transformed graphs for the embedding and decoder layers by using the popxl_addons.transforms.batch_serialisation.batch_serialise_fwd_and_grad directly. As for head layer that has a combined forward and gradient graph, it uses popxl_addons.transforms.batch_serialisation.batch_serialise.

For batch serialisation, we also need to create remote buffers to load the inputs and store outputs for each partition by using popxl_addons.batch_serial_buffer. In this application, we use the remote buffers x_buffer and dx_buffer respectively to handle the intermediate outputs of each partition in the forward pass and backward pass. The two buffers for this application are illustrated in the following diagram. Each row handles config.gradient_accumulation elements.

Buffers x_buffer and dx_buffer

For instance, in x_buffer, row 0 stores the output of the embedding layer in forward pass. The output of each GPT decoder layer is stored from row 1 to config.model.layers+1. Note that the rows in the two buffers are filled up in the opposite directions.

Execution of layers

Below are diagrams demonstrating how each layer is executed during the forward, backward and optimiser steps.

Forward layers

Forward layer execution and communication

Backwards layers

Backward layer execution and communication

Optimiser layers

Optimiser layer execution and communication

Note that the optimiser layer operates directly on the RTS sharded gradient accumulators, optimiser state and variables.