INDEX

Explanations

names of researchers and authors in cited studies

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

et al. citations

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

 ब्रेकडाउन

-0.47

AccessorTable

-0.44

 anún

-0.44

 læng

-0.42

 Pueden

-0.41

 senhora

-0.41

RenderAtEndOf

-0.41

dø

-0.40

 totalidad

-0.40

 praça

-0.40

POSITIVE LOGITS

 ImageButton

0.56

Identyfik

0.53

 Obrador

0.50

########.

0.48

et

0.48

 ImageIcon

0.46

BufferException

0.46

belea

0.46

 Merkel

0.45

ItemBackground

0.45

Activations Density 0.307%

names of researchers and authors in cited studies

et al. citations

No Comments

No Known Activations