INDEX

Explanations

actions related to removing or deleting elements or items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

textTheme

-0.59

 tantôt

-0.56

 defStyleAttr

-0.50

 fermés

-0.49

bagi

-0.49

 TextAlign

-0.48

 particulières

-0.47

ToUse

-0.46

splitting

-0.46

Leggi

-0.46

POSITIVE LOGITS

 unwanted

1.00

 unnecessary

0.98

 superfluous

0.89

 khỏi

0.88

 offending

0.88

掉

0.86

 excess

0.85

 redundant

0.84

 extraneous

0.83

Removes

0.83

Activations Density 0.333%

actions related to removing or deleting elements or items

No Comments

No Known Activations