zjr2000 commited on
Commit
cd4580d
·
verified ·
1 Parent(s): 7f3a591

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +0 -7
README.md CHANGED
@@ -19,10 +19,6 @@ SPES-2B is a pretrained language model released as part of paper:
19
 
20
  **Pretraining A Large Language Model using Distributed GPUs: A Memory-Efficient Decentralized Paradigm**
21
 
22
- ## Overview
23
-
24
- SPES-2B is a research checkpoint for studying decentralized and memory-efficient pretraining of Mixture-of-Experts (MoE) language models. The model is trained under the SPES framework, which enables distributed training across GPU nodes with lightweight expert synchronization.
25
-
26
  ## Model Details
27
 
28
  - **Model name:** SPES-2B
@@ -44,9 +40,6 @@ This model is intended for:
44
  - research on MoE training and synchronization
45
  - experimentation and evaluation of pretrained language models
46
 
47
- ## Notes
48
-
49
- This is a research model. Performance may vary depending on evaluation setup, prompting format, and downstream tasks.
50
 
51
  ## Citation
52
 
 
19
 
20
  **Pretraining A Large Language Model using Distributed GPUs: A Memory-Efficient Decentralized Paradigm**
21
 
 
 
 
 
22
  ## Model Details
23
 
24
  - **Model name:** SPES-2B
 
40
  - research on MoE training and synchronization
41
  - experimentation and evaluation of pretrained language models
42
 
 
 
 
43
 
44
  ## Citation
45