oopere/Llama-3.2-1B-pruned-20pct
Text Generation
•
1B
•
Updated
•
295
Llama 3.2 mlp pruned
Created by pruning the MLP (feedforward) layers, reducing the size of Llama models while improving their performance.
-
-
oopere/Llama-3.2-1B-pruned-40pct
Text Generation • 0.9B • Updated • 131 • 2 -
oopere/pruned60-llama-3.2-1B
Text Generation • 0.8B • Updated • 92 • 1 -
oopere/pruned40-llama-3.2-3b
Text Generation • 2B • Updated • 9 -
oopere/pruned20-llama-3.2-3b
Text Generation • 3B • Updated • 8 -
oopere/pruned60-llama-3.2-3b
Text Generation • 2B • Updated • 5 -
oopere/pruned10-llama-3.2-3B
Text Generation • 3B • Updated • 8