INDEX

Explanations

system error messages

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

syscall definitions

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

varlak

-0.38

lium

-0.37

UnusedPrivate

-0.34

MockBean

-0.34

ilkan

-0.33

 Momb

-0.32

dshaw

-0.32

oplasma

-0.32

 rusty

-0.32

omite

-0.32

POSITIVE LOGITS

 autorytatywna

0.52

ItemBackground

0.52

 kasarigan

0.51

 للمعارف

0.51

BeginInit

0.50

AndEndTag

0.50

 defaultstate

0.48

Gön

0.47

évaluateur

0.47

utilisons

0.46

Activations Density 0.077%

system error messages

syscall definitions

No Comments

No Known Activations