bigcode-project/Megatron-LM Issues
Without "<suffix>" token
Closed 1LM Head FLOPs
Updated 2Train Python model with FIM
Closed 4how to convert huggingface?
Closed 2Script to train starcoder
Closed 2TF-Model Architecture
Closed 2TF-Tokenization
ClosedExperiment plan
Closed 1Log GPU throughput
ClosedCreate the Stack 1.2 dataset
Closed 1Create data composition
Updated 1Multiple validation datasets
Closed 3Meta-information dropout
Closed 2Literature review on scaling laws
Updated 3Wandb init error
Updated