INDEX

Explanations

names of companies and brands

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

Capitalized words followed by a specific word

np_acts-logits-general · gemini-2.0-flash

company names and entities

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

-0.62

-0.54

all

-0.47

the

-0.47

con

-0.46

up

-0.46

out

-0.46

</h3>

-0.46

‍♀️

-0.44

-0.43

POSITIVE LOGITS

 protoimpl

1.04

 Monfieur

1.01

 Houſe

1.00

 архивлан

0.97

 Majefty

0.95

 Jefus

0.94

 Diſ

0.94

 purpoſe

0.93

 feroit

0.93

 houſe

0.93

Activations Density 0.988%

names of companies and brands

Capitalized words followed by a specific word

company names and entities

No Comments

No Known Activations