INDEX

Explanations

elements related to inquiry and knowledge acquisition

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

discovering the unknown

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

oa̍t

-0.51

riwal

-0.49

 Audiodateien

-0.49

brum

-0.48

InjectAttribute

-0.47

LElement

-0.47

chyma

-0.45

istration

-0.45

Хьажоргаш

-0.45

ficial

-0.44

POSITIVE LOGITS

 surprises

0.52

RectangleBorder

0.50

 surpresa

0.49

 Überras

0.46

 surprise

0.45

 blindly

0.45

ClCompile

0.41

 sorpresa

0.39

 expectations

0.39

 preview

0.38

Activations Density 0.055%

elements related to inquiry and knowledge acquisition

discovering the unknown

No Comments

No Known Activations