INDEX

Explanations

negative descriptors or phrases associated with poor quality or negative attributes

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

"bad"

np_acts-logits-general · gemini-2.0-flash

bad things and negative qualities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

 BoxFit

-0.70

 חיצוניים

-0.66

AxisAlignment

-0.64

GOTREF

-0.60

 Sumber

-0.58

日閲覧

-0.57

={({

-0.56

 createState

-0.56

testng

-0.54

\{\\

-0.54

POSITIVE LOGITS

bad

0.86

bad

0.78

 worse

0.75

Bad

0.74

Bad

0.72

BAD

0.67

CastException

0.66

 badly

0.65

 Worse

0.63

BAD

0.63

Activations Density 0.243%

negative descriptors or phrases associated with poor quality or negative attributes

"bad"

bad things and negative qualities

No Comments

No Known Activations