INDEX

Explanations

references to authority and moral judgment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

Efq

-0.79

 faſt

-0.69

Autoritní

-0.64

twimg

-0.61

互联网档案馆

-0.61

 fhort

-0.61

__':

-0.60

 autorytatywna

-0.59

 defire

-0.59

setVerticalGroup

-0.58

POSITIVE LOGITS

 >=",

0.58

INCREF

0.57

 empe

0.53

 eens

0.53

 ©️

0.50

SharedCtor

0.49

JSONException

0.49

offsetTop

0.49

ρέπει

0.47

 speciale

0.47

Activations Density 0.107%

references to authority and moral judgment

No Comments

No Known Activations