Those three parts consist of the encoder, the “decoder” (which actually consists of the decoder with the language modeling head), and the “decoder” with pre-computed key/values as additional inputs. This specific export comes from the fact that during the first pass, the decoder has no pre-computed…

What does the decoder with past values means

Rkoy July 7, 2025, 11:06am 3

1 Like

Topic		Replies	Views
Default for the Decoder past_key_values - Marian Intermediate	0	409	January 5, 2023
The way to get Seq2SeqLM's `decoder_input_ids` to obtain `past_key_values` Beginners	0	1358	October 25, 2020
Why past_key_values is not in GreedySearchDecoderOnlyOutput? 🤗Transformers	1	2047	October 4, 2022
Control EncoderDecoderModel to generate tokens step by step 🤗Transformers	8	2607	June 8, 2022
Past_key_values - why not past_key_values_queries? Beginners	5	11183	October 15, 2023