bigscience

Research workshop on large language models - The Summer of Language Models 21

At the moment we have 2 code repos:

1. https://github.com/bigscience-workshop/Megatron-DeepSpeed - this is our flagship code base
2. https://github.com/bigscience-workshop/bigscience - (this repo) for everything else - docs, experiments, etc.

Currently, the most active segments of this repo are:

• JZ - Lots of information about our work environment which helps evaluate, plan and get things done
• Experiments - many experiments are being done. Documentation, result tables, scripts and logs are all there
• Datasets info
• Train - all the information about the current trainings (see below for the most important ones)

We have READMEs for specific aspects, such as: - hub integration

Trainings

While we keep detailed chronicles of experiments and findings for some of the main trainings, here is a doc that contains a summary of the most important findings: Lessons learned

Train 1 - 13B - unmodified Megatron gpt2 - baseline

• the full spec and discussions
• the training script
• checkpoints and logs:
  • tensorboard
  • logs
• chronicles

You can watch the training logs live by running this tail -f like script over remote log file that gets synced to the hub once an hour: ``` perl -e '$u=shift; $b=0; while(1){($e)=qx[curl -sI $u]=~/content-length: (\d+)/; \ print qx[curl -sr $b-$e -L $u] if $e>$b; $b=$e; sleep 300}' \ https://huggingface.co/bigscience/tr1-13B-logs/resolve/main/main_log.txt

```

Train 3

Architecture and scaling baseline runs: no fancy tricks, just GPT2. Here are links to the respective tensorboards:

| Size | 1B3 | 760M | 350M | 125M | |--------------------- |----- |------ |------ |------ | | C4 + low warmup | a | b | c | | | OSCAR + low warmup | f | | | | | C4 + high warmup | e | | | | | OSCAR + high warmup | d (current baseline) | g | h | i | | Pile + high warmup | m | j | k | l |

Train 8

104B - unmodified Megatron gpt2 - with extra-wide hidden size to learn how to deal with training instabilities

• the full spec and discussions
• the training script
• checkpoints and logs:
  • tensorboard
  • logs
• chronicles

You can watch the training logs live by running this tail -f like script over remote log file that gets synced to the hub once an hour: perl -e '$u=shift; $b=0; while(1){($e)=qx[curl -sI $u]=~/content-length: (\d+)/; \ print qx[curl -sr $b-$e -L $u] if $e>$b; $b=$e; sleep 300}' \ https://cdn-lfs.huggingface.co/bigscience/tr8-104B-logs/b2cc478d5ae7c9ec937ea2db1d2fe09de593fa2ec38c171d6cc5dca094cd79f9

Train 11

This is the current main training

tr11-176B-ml

• the full spec and discussions
• the training script
• checkpoints and logs:
  • tensorboard
  • logs
• chronicles-prequel
• chronicles

You can watch the training logs live by running this tail -f like script over remote log file that gets synced to the hub once an hour: perl -e '$u=shift; $b=0; while(1){($e)=qx[curl -LsI $u]=~/2 200.*?content-length: (\d+)/s; \ print qx[curl -Lsr $b-$e $u] if $e>$b; $b=$e; sleep 300}' \ https://huggingface.co/bigscience/tr11-176B-ml-logs/resolve/main/logs/main/main_log.txt