Merge branch 'github-pr' into 'main'

Combination of several github PRs

See merge request ADLR/megatron-lm!383

README.md

-Megatron ([1](https://arxiv.org/pdf/1909.08053.pdf) and [2](https://arxiv.org/pdf/2104.04473.pdf)) is a large, powerful transformer developed by the Applied Deep Learning Research team at NVIDIA. This repository is for ongoing research on training large transformer language models at scale. We developed efficient, model-parallel (tensor and pipeline), and multi-node pre-training oftransformer based models such as [GPT](https://arxiv.org/abs/2005.14165), [BERT](https://arxiv.org/pdf/1810.04805.pdf), and [T5](https://arxiv.org/abs/1910.10683) using mixed precision.
+Megatron ([1](https://arxiv.org/pdf/1909.08053.pdf) and [2](https://arxiv.org/pdf/2104.04473.pdf)) is a large, powerful transformer developed by the Applied Deep Learning Research team at NVIDIA. This repository is for ongoing research on training large transformer language models at scale. We developed efficient, model-parallel (tensor and pipeline), and multi-node pre-training of transformer based models such as [GPT](https://arxiv.org/abs/2005.14165), [BERT](https://arxiv.org/pdf/1810.04805.pdf), and [T5](https://arxiv.org/abs/1910.10683) using mixed precision.
 
 Below are some of the projects where we have directly used Megatron:
 * [BERT and GPT Studies Using Megatron](https://arxiv.org/pdf/1909.08053.pdf)
@@ -13,7 +13,9 @@ Below are some of the projects where we have directly used Megatron:
 
 Megatron is also used in [NeMo Megatron](https://developer.nvidia.com/nvidia-nemo#nemo-megatron), a framework to help enterprises overcome the challenges of building and training sophisticated natural language processing models with billions and trillions of parameters.
 
-Our codebase is capable of efficiently training very large (hundreds of billions of parameters) language models with both model and data parallelism. To demonstrate how the code scales with multiple GPUs and model sizes, we consider GPT models from 1 billion all the way to 1 trillion parameters. All models use a vocabulary size of 51,200 and a sequence length of 2048. We vary hidden size, number of attention heads, and number of layers to arrive at a specifc model size. As the model size increases, we also modestly increase the batch size. We leverage [NVIDIA's Selene supercomputer](https://www.top500.org/system/179842/) to perform scaling studies and use up to 3072 [A100](https://www.nvidia.com/en-us/data-center/a100/) GPUs for the largest model. The table below shows the model configurations along with the achieved FLOPs (both per GPU and aggregate over all GPUs). Note that these results are from benchmark runs and these models were not trained to convergence; however, the FLOPs are measured for end-to-end training, i.e., includes all operations including data loading, optimization, and even logging.
+Our codebase is capable of efficiently training very large (hundreds of billions of parameters) language models with both model and data parallelism. To demonstrate how the code scales with multiple GPUs and model sizes, we consider GPT models from 1 billion all the way to 1 trillion parameters. All models use a vocabulary size of 51,200 and a sequence length of 2048. We vary hidden size, number of attention heads, and number of layers to arrive at a specifc model size. As the model size increases, we also modestly increase the batch size. We leverage [NVIDIA's Selene supercomputer](https://www.top500.org/system/179842/) to perform scaling studies and use up to 3072 [A100](https://www.nvidia.com/en-us/data-center/a100/) GPUs for the largest model. Each cluster node has 8 NVIDIA 80GB A100 GPUs. The table below shows the model configurations along with the achieved FLOPs (both per GPU and aggregate over all GPUs). Note that these results are from benchmark runs and these models were not trained to convergence; however, the FLOPs are measured for end-to-end training, i.e., includes all operations including data loading, optimization, and even logging.
+
+Additionally, the model parallel size column reports a combined tensor and pipeline parallelism degrees. For numbers larger than 8, typically tensor parallel of size 8 was used. So, for example, the 145B model reports the total model parallel size of 64, which means that this setup used TP=8 and PP=8.
 
 ![Cases](images/cases_april2021.png)
 
@@ -29,7 +31,6 @@ All the cases from 1 billion to 1 trillion parameters achieve more than 43% half
       * [Data Preprocessing](#data-preprocessing)
       * [BERT Pretraining](#bert-pretraining)
       * [GPT Pretraining](#gpt-pretraining)
-      * [GPT Pretraining](#gpt-pretraining)
       * [T5 Pretraining](#t5-pretraining)
       * [Distributed Pretraining](#distributed-pretraining)
       * [GPT-3 Example](#gpt-3-example)
@@ -206,7 +207,7 @@ Further command line arguments are described in the source file [`arguments.py`]
 
 ## T5 Pretraining
 
-Very similar to BERT and GPT, the `examples/pretrain_t5.sh` script runs single GPU "base" (~220M parameter) T5 pretraining. The primary difference from BERT and GPT is the addition of the following arguments to accomodate the T5 architecture:
+Very similar to BERT and GPT, the `examples/pretrain_t5.sh` script runs single GPU "base" (~220M parameter) T5 pretraining. The primary difference from BERT and GPT is the addition of the following arguments to accommodate the T5 architecture:
 
 * `--kv-channels` sets the inner dimension of the "key" and "value" matrices of all attention mechanisms in the model. For BERT and GPT this defaults to the hidden size divided by the number of attention heads, but can be configured for T5.
 
@@ -262,7 +263,7 @@ Second, we developed a simple and efficient two-dimensional model-parallel appro
 
 <!-- The number of microbatches in a per-pipeline minibatch is controlled by the `--num-microbatches-in-minibatch` argument. With `WORLD_SIZE` GPUs, `TENSOR_MP_SIZE` tensor-model-parallel size, `PIPELINE_MP_SIZE` pipeline-model-parallel-size, `WORLD_SIZE`/(`TENSOR_MP_SIZE` * `PIPELINE_MP_SIZE`) GPUs will be used for data parallelism. The default values for `--tensor-model-parallel-size` and `--pipeline-model-parallel-size` is 1, which will not implement either form of model parallelism. -->
 
-We have examples of how to use these two different forms of model parallelism the example scripts ending in `distributed_with_mp.sh`, note that pipeline parallelism is not currently supported in the T5 model:
+We have examples of how to use these two different forms of model parallelism the example scripts ending in `distributed_with_mp.sh`:
 
 Other than these minor changes, the distributed training is identical to the training on a single GPU.
 
@@ -399,7 +400,7 @@ python tools/create_doc_index.py \
 
 We provide several command line arguments, detailed in the scripts listed below, to handle various zero-shot and fine-tuned downstream tasks. However, you can also finetune your model from a pretrained checkpoint on other corpora as desired. To do so, simply add the `--finetune` flag and adjust the input files and training parameters within the original training script. The iteration count will be reset to zero, and the optimizer and internal state will be reinitialized. If the fine-tuning is interrupted for any reason, be sure to remove the `--finetune` flag before continuing, otherwise the training will start again from the beginning.
 
-Because evaluation requires substantially less memory than training, it may be advantageous to merge a model trained in parallel for use on a single GPU in downstream tasks. The following script accomplishes this. Currently only tensor model parallelism is supported on input and pipeline model parallelsim on the output. This example reads in a model with 2-way tensor model parallelism and writes out a model with 2-way pipeline model parallelism.
+Because evaluation requires substantially less memory than training, it may be advantageous to merge a model trained in parallel for use on a single GPU in downstream tasks. The following script accomplishes this. Currently only tensor model parallelism is supported on input and pipeline model parallelism on the output. This example reads in a model with 2-way tensor model parallelism and writes out a model with 2-way pipeline model parallelism.
 
 <pre>
 TENSOR_MODEL_PARALLEL_SIZE=2
@@ -484,7 +485,7 @@ python tasks/main.py \
 
 
 ### LAMBADA Cloze Accuracy
-To compute LAMBADA cloze accuracy (the accuracy of predicting the last token given the preceeding tokens) we utilize a detokenized, processed version of the [LAMBADA dataset](https://github.com/cybertronai/bflm/blob/master/lambada_test.jsonl).
+To compute LAMBADA cloze accuracy (the accuracy of predicting the last token given the preceding tokens) we utilize a detokenized, processed version of the [LAMBADA dataset](https://github.com/cybertronai/bflm/blob/master/lambada_test.jsonl).
 
 We use the following command to run LAMBADA evaluation on a 345M parameter model. Note that the `--strict-lambada` flag should be used to require whole word matching. Make that `lambada` is part of the file path.
 

examples/msdp/prep_resp_gen.sh

@@ -14,5 +14,5 @@ PROCESSED_FILE=<PATH_OF_INPUT_FILE_FOR_RESPONSE_GENERATION> \
 python ${DIR}/tasks/msdp/preprocessing.py \
         --func prepare_input \
         --test_file ${TEST_FILE} \
-        --knowledge_gen_file ${KNOWLEDGE_FILE} \
+        --knwl_gen_file ${KNOWLEDGE_FILE} \
         --processed_file ${PROCESSED_FILE}

megatron/mpu/layers.py

@@ -240,7 +240,7 @@ class ColumnParallelLinear(torch.nn.Module):
         input_size: first dimension of matrix A.
         output_size: second dimension of matrix A.
         bias: If true, add bias
-        gather_output: If true, call all-gether on output and make Y avaiable
+        gather_output: If true, call all-gather on output and make Y avaiable
                        to all GPUs, otherwise, every GPU will have its output
                        which is Y_i = XA_i
         init_method: method to initialize weights. Note that bias is always set