Skip to main content

Mantis-8B-CLIP-LLaMA-3

This is the training curves of model Mantis-8B-CLIP-LLaMA-3. It's reported to help reproduce mantis's training results. This training has 2 training failures, and models trainings are resumed from the previous checkpoint with the same hyper parameters
Created on August 5|Last edited on August 5


Section 1


1k2k3k4k5k6ktrain/global_step00.0000020.0000040.0000060.0000080.00001
1k2k3k4k5k6ktrain/global_step50100150200250
1k2k3k4k5k6ktrain/global_step1234
Run set
10



Run set
10