Method encode

llms/speculative_decoding/model.py:292–293 · view source on GitHub ↗

(self, inputs: mx.array)

Source from the content-addressed store, hash-verified

290	self.reset_cache()
291
292	def encode(self, inputs: mx.array):
293	return self.encoder(self.wte(inputs))
294
295	def truncate_cache(self, num_to_truncate):
296	if num_to_truncate <= 0:

__call__Method · 0.95

no outgoing calls

no test coverage detected