INDEX

Explanations

phrases indicating positions or moments in time, particularly in the context of sports or events

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

immediately followed by "in" or "that"

np_acts-logits-general · gemini-2.0-flash

inthroughoutchildWheel

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

/@

-0.54

INFIN

-0.52

 PLWABN

-0.50

ence

-0.50

 moment

-0.50

astify

-0.50

 Himself

-0.49

 updatedAt

-0.49

 ourselves

-0.49

 Fortun

-0.48

POSITIVE LOGITS

tinyos

0.66

 lenker

0.63

jspb

0.60

 bezeichneter

0.57

 esternos

0.57

expandindo

0.56

reportWebVitals

0.56

UAGES

0.55

RTHOOK

0.54

UnknownFields

0.53

Activations Density 0.130%

phrases indicating positions or moments in time, particularly in the context of sports or events

immediately followed by "in" or "that"

inthroughoutchildWheel

No Comments

No Known Activations