Pre-trained models in MiniPLM: Knowledge Distillation for Pre-Training Language Models
MiniLLM
community
AI & ML interests
Training efficient language models (MiniLLM, MiniPLM)
Recent Activity
View all activity
Organization Card
Collections
2
models
50
MiniLLM/init-gpt2-120M
Text Generation
•
Updated
•
346
•
1
MiniLLM/teacher-Llama-13B
Text Generation
•
Updated
•
6
MiniLLM/MiniLLM-Llama-7B
Text Generation
•
Updated
•
10
MiniLLM/Ref-Pretrain-Qwen-104M
Text Generation
•
Updated
•
28
•
1
MiniLLM/MiniPLM-Mamba-130M
Text Generation
•
Updated
•
33
•
1
MiniLLM/MiniPLM-Qwen-1.2B
Text Generation
•
Updated
•
37
•
2
MiniLLM/MiniPLM-Qwen-500M
Text Generation
•
Updated
•
439
•
5
MiniLLM/MiniPLM-Qwen-200M
Text Generation
•
Updated
•
447
MiniLLM/MiniPLM-llama3.1-212M
Text Generation
•
Updated
•
29
•
1
MiniLLM/Pretrain-Qwen-500M
Text Generation
•
Updated
•
24
datasets
10
MiniLLM/pile-tokenized
Updated
•
69
MiniLLM/roberta-corpus-processed
Updated
•
55
MiniLLM/pile-diff_samp-qwen_1.8B-qwen_104M-r0.5
Updated
•
72
MiniLLM/openwebtext-processed
Updated
•
70
MiniLLM/dolly-processed
Viewer
•
Updated
•
110k
•
306
•
1
MiniLLM/sinst
Viewer
•
Updated
•
8.35k
•
64
•
1
MiniLLM/uinst
Viewer
•
Updated
•
64.8k
•
124
•
1
MiniLLM/self-inst
Viewer
•
Updated
•
242
•
64
•
2
MiniLLM/Vicuna
Viewer
•
Updated
•
80
•
65
•
1
MiniLLM/dolly
Viewer
•
Updated
•
500
•
104