deepseek-ai/ESFT-vanilla-lite
Text Generation
•
16B
•
Updated
•
551
•
19
ESFT
models for paper expert-specialized fine-tuning
-
-
deepseek-ai/ESFT-token-law-lite
Text Generation • 1B • Updated • 57 • 6 -
deepseek-ai/ESFT-token-summary-lite
Text Generation • 1B • Updated • 49 • 4 -
deepseek-ai/ESFT-token-code-lite
Text Generation • 2B • Updated • 48 • 4 -
deepseek-ai/ESFT-token-intent-lite
Text Generation • 2B • Updated • 53 • 3 -
deepseek-ai/ESFT-token-translation-lite
Text Generation • 0.9B • Updated • 71 • 3 -
deepseek-ai/ESFT-token-math-lite
Text Generation • 1B • Updated • 52 • 4 -
deepseek-ai/ESFT-gate-law-lite
Text Generation • 2B • Updated • 55 • 4 -
deepseek-ai/ESFT-gate-summary-lite
Text Generation • 2B • Updated • 56 • 3 -
deepseek-ai/ESFT-gate-translation-lite
Text Generation • 1B • Updated • 53 • 4 -
deepseek-ai/ESFT-gate-code-lite
Text Generation • 2B • Updated • 57 • 3 -
deepseek-ai/ESFT-gate-intent-lite
Text Generation • 2B • Updated • 57 • 4 -
deepseek-ai/ESFT-gate-math-lite
Text Generation • 2B • Updated • 59 • 3 -
Let the Expert Stick to His Last: Expert-Specialized Fine-Tuning for Sparse Architectural Large Language Models
Paper • 2407.01906 • Published • 46