INDEX

Explanations

code snippets related to user data management and UI actions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

Around the word "edit" or editing-related terms

np_acts-logits-general · gemini-2.0-flash

edit or create

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

الحياه

-0.54

녁

-0.48

Према

-0.43

 Cosa

-0.42

SetString

-0.41

 CreateTagHelper

-0.40

 conveyed

-0.38

 đừng

-0.37

言葉

-0.37

 Secondo

-0.37

POSITIVE LOGITS

 edit

1.49

 Edit

1.38

 editing

1.28

edit

1.23

Edit

1.18

 Editing

1.08

 EDIT

1.06

编辑

1.04

 edits

1.01

 edited

0.96

Activations Density 0.263%

code snippets related to user data management and UI actions

Around the word "edit" or editing-related terms

edit or create

No Comments

No Known Activations