DistilLED Large CNN 16384
distil-led-large-cnn-16384 was initialized from sshleifer/distilbart-cnn-12-6, in a fashion similar to allenai/led-large-16384.
To be able to process 16K tokens, sshleifer/distilbart-cnn-12-6's position embedding matrix was simply copied 16 times.
This checkpoint should be loaded into LEDForConditionalGeneration.from_pretrained. See the LED documentation for more information.
- Downloads last month
- 235
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social
visibility and check back later, or deploy to Inference Endpoints (dedicated)
instead.