INDEX

Explanations

instances of the word "to" and its related forms in various contexts

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

Comes before verbs

np_acts-logits-general · gemini-2.0-flash

going to places or actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

verwijspagina

-0.71

 lenker

-0.60

ագրություններ

-0.60

 Josephus

-0.58

 CURIAM

-0.58

 Morav

-0.55

 Whigs

-0.55

 Tasmanian

-0.54

 Cebuano

-0.54

 Volga

-0.54

POSITIVE LOGITS

the

0.76

 places

0.73

ReusableCell

0.65

see

0.65

arsis

0.63

bed

0.60

TagHelper

0.58

 destinations

0.58

بوابة

0.58

 visit

0.57

Activations Density 0.208%

instances of the word "to" and its related forms in various contexts

Comes before verbs

going to places or actions

No Comments

No Known Activations