INDEX

Explanations

legal terminology and procedural language

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

any further offenses

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

 <<<<<<<<<<<<<<

-0.50

OGND

-0.46

Havolalar

-0.43

ьажоргаш

-0.41

 jScrollPane

-0.40

 AssemblyCulture

-0.38

 overrated

-0.36

 المعيارى

-0.36

Jir

-0.35

ernoo

-0.35

POSITIVE LOGITS

tagHelperRunner

0.56

 kasarigan

0.52

 dezelve

0.44

AndEndTag

0.44

genodigd

0.42

lighed

0.42

eleste

0.41



0.41

 beſte

0.41

 dieſes

0.41

Activations Density 0.004%

legal terminology and procedural language

any further offenses

No Comments

No Known Activations

legal terminology and procedural language

any further offenses

No Comments

No Known Activations