# Obtain step by step outputs of model.generate

**URL:** <https://discuss.huggingface.co/t/obtain-step-by-step-outputs-of-model-generate/19709>\
**Category:** 🤗Transformers\
**Created:** [June 28, 2022, 7:15am UTC](https://discuss.huggingface.co/t/obtain-step-by-step-outputs-of-model-generate/19709 "2022-06-28T07:15:58Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![kumarashutosh](https://avatars.discourse-cdn.com/v4/letter/k/94ad74/32.png) [@kumarashutosh](https://discuss.huggingface.co/u/kumarashutosh)\
**Post date:** [June 28, 2022, 7:15am UTC](https://discuss.huggingface.co/t/obtain-step-by-step-outputs-of-model-generate/19709/1 "2022-06-28T07:15:58Z")

</div>

Hi,  
I am using `LEDForConditionalGeneration` for fine-tuning a summarization model. I am able to finetune the network and obtain good results for my purpose. But now, I want to modify the encoder to use some different architecture. How can I break the `model.generate()` into individual components of the transformer?  
For example, I want something like this:

```auto
outputs = model(input_ids)
encoder_outputs = outputs["encoder_last_hidden_state"] #For encoder outputs
# Similar logic for decoder outputs and then beam search to get text sentences

```

How can I obtain such encoder decoder outputs and obtain exactly the same result that I get with `model.generate`? A sample code for `model.generate` is as follows

```auto
ARTICLE_TO_SUMMARIZE = "Some long document that I want to summarize.
inputs = tokenizer.encode(ARTICLE_TO_SUMMARIZE, return_tensors="pt")
global_attention_mask = torch.zeros_like(inputs)
global_attention_mask[:, 0] = 1
# Generate Summary
summary_ids = model.generate(inputs, global_attention_mask=global_attention_mask, num_beams=3, max_length=32)
print(tokenizer.decode(summary_ids[0], skip_special_tokens=True, clean_up_tokenization_spaces=False))

```

Any help in this direction will be appreciated. Thanks!