INDEX

Explanations

conjunctions and phrases indicating addition or inclusion

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

and the

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

зий

-0.41

mimeType

-0.40

repr

-0.38

แห่ง

-0.36

Kr

-0.36

CrossRef

-0.35

MOCK

-0.34

vode

-0.32

 menetap

-0.32

Kre

-0.32

POSITIVE LOGITS

 initComponents

0.73

RegistryLite

0.61

 wireType

0.56

writeFieldEnd

0.54

 Chwiliwch

0.54

 ویکی‌پدی

0.54

Хьажоргаш

0.50

 Савезне

0.50

↹

0.49

хьтан

0.49

Activations Density 0.051%

conjunctions and phrases indicating addition or inclusion

and the

No Comments

No Known Activations