SPAM: Spike-Aware Adam with Momentum Reset for Stable LLM Training
Paper
•
2501.06842
•
Published
•
16
Shuyi Li
WoXiangSi
·
AI & ML interests
None yet
Recent Activity
liked
a dataset
8 days ago
yaodi/shijing
liked
a model
16 days ago
mistralai/Mistral-7B-Instruct-v0.3
liked
a model
17 days ago
meta-llama/Meta-Llama-3-8B-Instruct
Organizations
