INDEX

Explanations

expressions of disappointment or frustration regarding experiences

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

asking questions or giving instructions

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

Hochspringen

-0.56

 Administrativna

-0.54

 betweenstory

-0.48

𝟹

-0.46

 continúas

-0.46

Архівовано

-0.45

 nonUne

-0.45

 kohdetta

-0.42

目

-0.41

flüge

-0.41

POSITIVE LOGITS

исленность

0.45

 pinulongan

0.40

 nahilalakip

0.40

Trama

0.37

Filmografia

0.36

Gön

0.35

AndroidJUnit

0.35

NewUrlParser

0.34

 Normdatei

0.34

 lenker

0.33

Activations Density 0.020%

expressions of disappointment or frustration regarding experiences

asking questions or giving instructions

No Comments

No Known Activations