how to get the hidden features by api code?

[hardlipay]

encoder_outputs = self.encoder(
            embedding_output,
            head_mask=head_mask,
            output_attentions=output_attentions,
            output_hidden_states=output_hidden_states,
            return_dict=return_dict,
        )

Here is the output returned by dinov2 in hg/transformer. Just encoder_outputs.output_hidden_states to get the [bs, 257, 768] feature tensor.

I am now using the openclip library for the eva02 model and I can't find a similar return. Please let me know what code I should use, thanks! 这里是dinov2在hg/transformer中返回的输出。 只要encoder_outputs.output_hidden_states就可以获取【bs，257，768】的特征张量。

我现在在使用openclip库的eva02模型，我找不到类似的返回。 请告诉我应该使用什么样的代码，谢谢

[hardlipay]

    def forward(self, x):
        x = self.forward_features(x)
        x = self.forward_head(x)
        return x

    [](https://github.com/huggingface/pytorch-image-models/blob/0bc9a63ee5e432c85c8f87d7b116a0430ea3764a/timm/models/eva.py#L579C3-L582C17)

    just to

    ```
def forward(self, x):
    x_h = self.forward_features(x) # right?
    x = self.forward_head(x_h)
    return x,x_h