INDEX

Explanations

phrases indicating investigation or analysis in scientific research contexts

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

research and experiments

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

#+#

-0.56

jsxFileName

-0.52

 utafitiHapana

-0.49

RRect

-0.48

 dAtA

-0.46

 المعيارى

-0.46

Toole

-0.45

Kanpo

-0.45

StatusLabel

-0.44

culable

-0.43

POSITIVE LOGITS

wußt

0.48

 StatelessWidget

0.48

 Савезне

0.44

enumii

0.40

 beroep

0.38

ScopeManager

0.36

IsMutable

0.35

addPreferredGap

0.35

 experimentally

0.34

 initComponents

0.34

Activations Density 0.168%

phrases indicating investigation or analysis in scientific research contexts

research and experiments

No Comments

No Known Activations