Eric Harper
Eric Harper
# What does this PR do ? Adds rampup batch size to NeMo Megatron GPT. Scales the batch size linearly from starting batch size to global batch size by a...
# What does this PR do ? Add a one line overview of what this PR aims to accomplish. **Collection**: [Note which collection this PR will affect] # Changelog -...
# What does this PR do ? Add a one line overview of what this PR aims to accomplish. **Collection**: [Note which collection this PR will affect] # Changelog -...