INDEX

Explanations

sentences that contain punctuation marks or notable sentence breaks

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

numbers, adjectives, or compounds

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

Personendaten

-0.71

webElementXpaths

-0.68

DotNetBar

-0.62

jsxFileName

-0.59

 Chwiliwch

-0.59

ſammen

-0.58

 ब्रेकडाउन

-0.57

Personensuche

-0.55

 lizenzfreie

-0.55

########.

-0.54

POSITIVE LOGITS

rån

0.34

 voer

0.32

kę

0.30

 Duggan

0.30

 Schülerinnen

0.29

 sahiptir

0.29

 ước

0.29

köy

0.28

ніципа

0.28

diyesi

0.28

Activations Density 0.063%

sentences that contain punctuation marks or notable sentence breaks

numbers, adjectives, or compounds

No Comments

No Known Activations