Skip to main content
kastan
Log in
Sign up
Kastan Day
kastan
University of Illinois at Urbana-Champaign
kastanday
Teams
guangchen-rescale
Profile
Stars
Reports
Speed-Testing Colossal AI's Parallelism
A need for speed...
690 views
Last edit 3 years ago
AMP vs Custom Quantization
Accelerating multi-node Large Language Model training with per-layer selective quantization (e.g. FP32 -> FP16) of the transformer architecture.
464 views
Last edit 2 years ago
Activity
Mon
Wed
Fri
Nov
Dec
Jan
Feb
Mar
Apr
May
Jun
Jul
Aug
Sep
Oct
Runs
1-10
of 24
Name
Project
State
Created
spring-voice-21
huggingface
Crashed
2 years ago
feasible-planet-20
huggingface
Failed
2 years ago
honest-pyramid-19
huggingface
Failed
2 years ago
lucky-eon-18
huggingface
Killed
2 years ago
laced-moon-17
huggingface
Failed
2 years ago
decent-haze-16
huggingface
Failed
2 years ago
genial-night-15
huggingface
Failed
2 years ago
upbeat-cosmos-14
huggingface
Crashed
2 years ago
sf_medium_jul_13_1_epoch_bf16_8bit_bs4
huggingface
Killed
2 years ago
young-feather-12
huggingface
Killed
2 years ago
Loading...
