INDEX

Explanations

references to direction or distance in a journey

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

phrases that indicate direction or movement

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

all the way to

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

 quæ

-0.69

 wireType

-0.66

InstrumentedTest

-0.61

 suficientes

-0.60

 tubercle

-0.60

jstor

-0.58

PullParser

-0.58

complexType

-0.56

enschappelijke

-0.56

 mijne

-0.55

POSITIVE LOGITS

 down

0.61

up

0.59

 Мексичка

0.57

 till

0.56

antenimiento

0.55

 until

0.54

hamdu

0.54

 through

0.52

writeFieldEnd

0.49

 Bioaccumulative

0.49

Activations Density 0.006%

references to direction or distance in a journey

phrases that indicate direction or movement

all the way to

No Comments

No Known Activations