INDEX

Explanations

greetings and polite expressions

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

Greetings or welcomes

np_acts-logits-general · gemini-2.0-flash

greetings and introductions

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

saraba

-0.82

 AssemblyTitle

-0.68

 kasarigan

-0.67

PreferredItem

-0.66

ThroughAttribute

-0.65

indd

-0.63

 صوتيه

-0.63

ITHUB

-0.62

":[{

-0.62

Brainz

-0.62

POSITIVE LOGITS

 glad

0.63

 hope

0.60

Here

0.59

 👋

0.59

 welcome

0.58

 voici

0.58

 here

0.58

Firstly

0.57

 sorry

0.57

 Here

0.56

Activations Density 0.075%

greetings and polite expressions

Greetings or welcomes

greetings and introductions

No Comments

No Known Activations