INDEX

Explanations

instances of the word "is" and its related forms across varying contexts

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

which is

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

AntiForgeryToken

-0.42

AnchorTagHelper

-0.40

 Hofmann

-0.40

nthetic

-0.38

riwal

-0.35

hac

-0.35

 渋谷

-0.34

spli

-0.34

sett

-0.32

msgstr

-0.31

POSITIVE LOGITS

SharedDtor

0.82

NameInMap

0.59

 reaſon

0.59

 respeito

0.52

principalTable

0.51

Vidite

0.50

ViewFeatures

0.50

 kasarigan

0.49

 zelve

0.49

ſelf

0.48

Activations Density 0.040%

instances of the word "is" and its related forms across varying contexts

which is

No Comments

No Known Activations