INDEX

Explanations

phrases related to processes involving transformation and adaptation

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

immediate, every year, cross-sectional

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

锈钢

-0.48

complexContent

-0.47

 "..\..\..\

-0.43

 silent

-0.40

 estekak

-0.39

 "..\..\

-0.39

 ralenti

-0.37

+#+#

-0.37

 Signalez

-0.36

 ostavi

-0.36

POSITIVE LOGITS

 ویکی‌پدیا

0.59

fresh

0.52

immediate

0.50

 immedi

0.48

verwijspagina

0.46

Fresh

0.46

quick

0.46

 giornata

0.45

 Fresh

0.45

transient

0.44

Activations Density 0.155%

phrases related to processes involving transformation and adaptation

immediate, every year, cross-sectional

No Comments

No Known Activations