INDEX

Explanations

instances of the word "think" in various contexts

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

expressions of belief or opinion

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

i think

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

Efq

-1.27

ſelf

-1.21

ſelves

-1.15

 Majefty

-1.13

 Jefus

-1.11

 itſelf

-1.08

 himſelf

-1.07

 Reſ

-1.06

 Houſe

-1.03

 Diſ

-1.01

POSITIVE LOGITS

it

1.15

we

1.09

he

0.86

you

0.86

 they

0.85

0.84

 maybe

0.77

 this

0.75

the

0.72

 everyone

0.72

Activations Density 0.063%

instances of the word "think" in various contexts

expressions of belief or opinion

i think

No Comments

No Known Activations