INDEX

Explanations

instances of the word "like" and its variations, indicating a focus on preference or social engagement

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

Like this:

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_32/width_131k/average_l0_88

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.32.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ValueStyle

-0.64

 propOrder

-0.64

AndEndTag

-0.63

 فريبيس

-0.62

complexContent

-0.61

GraphicsUnit

-0.61

__':

-0.61

EDEFAULT

-0.60

saraba

-0.60

__":

-0.60

POSITIVE LOGITS

us

0.47

 Like

0.46

 LIKE

0.43

Liked

0.42

 Facebook

0.40

comment

0.40

 SHARE

0.40

 what

0.38

Share

0.38

 COMMENT

0.38

Activations Density 0.004%

instances of the word "like" and its variations, indicating a focus on preference or social engagement

Like this:

No Comments

No Known Activations