INDEX

Explanations

references to social or interpersonal dynamics involving respect and criticism

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

saying, respect, harsh, pun, against

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

setViewportView

-0.67

basicConfig

-0.54

 initComponents

-0.53

matchCondition

-0.49

 estekak

-0.49

 Füßen

-0.49

 createSlice

-0.48

 StatefulWidget

-0.48

outState

-0.48

 ervan

-0.48

POSITIVE LOGITS

$_['

0.36

 words

0.31

openzeppelin

0.31

 anyone

0.31

:+:

0.30

oth

0.30

btw

0.29

aop

0.29

vasti

0.28

fo

0.28

Activations Density 0.016%

references to social or interpersonal dynamics involving respect and criticism

saying, respect, harsh, pun, against

No Comments

No Known Activations