INDEX

Explanations

conditional statements or phrases indicating conditions or hypotheticals

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

if statements

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

noted

-0.42

 plumas

-0.41

 coats

-0.41

already

-0.40

Wad

-0.39

 coat

-0.39

 Universitaria

-0.39

note

-0.39

 Sache

-0.39

 Gode

-0.38

POSITIVE LOGITS

archiviato

0.54

Cyfarwyddwr

0.52

ьаж

0.50

jooq

0.48

رشف

0.48

kháu

0.47

stateProvider

0.45

੝

0.45

routeProvider

0.45

Kjelder

0.44

Activations Density 0.032%

conditional statements or phrases indicating conditions or hypotheticals

if statements

No Comments

No Known Activations