LoreonLabsPlatform

Overview

Intelligence

Markets
Builders
Research
Narratives
Ecosystems
Launchpads

Discover

Search
Sources

Python

Megatron-LM

Ongoing research training transformer models at scale

PythonEmerging

Stars

—

Forks

—

Contributors

8

Last push

3mo ago

Recent commits

Latest commits.

Merge branch 'decoder_input' into 'main'
cbc89b3Oliver Koenig15mo ago
ADLR/megatron-lm!2967 - Remove obsolete reference to decoder_input
530092bKeshav Santhanam15mo ago
Merge branch 'helenn-evals-completions-api' into 'main'
e5d3bf7Deepak Narayanan15mo ago
ADLR/megatron-lm!2959 - Re-enable completions endpoint
74f8865Helen Ngo15mo ago
Merge branch 'helenn-bert-failure-fix' into 'main'
876ff5aOliver Koenig15mo ago

ADLR/megatron-lm!2994 - Fix BERT signature in legacy path

32ea646Helen Ngo15mo ago

Merge branch 'shifang/moe_with_calculate_per_token_loss' into 'main'

6dfb5c2Oliver Koenig15mo ago

ADLR/megatron-lm!2851 - fix(moe): Fix MoE Aux loss scaling when calculate_per_token_loss=True

87d9d25Shifang Xu15mo ago

Top contributors

Builders behind this project.