INDEX

Explanations

terms that evaluate the practicality and appropriateness of ideas or proposals

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

possibility or suitability

np_acts-logits-general · gemini-2.0-flash

suitability and value

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

 ModelExpression

-0.92

expandindo

-0.83

 Мексичка

-0.82

 Administrativna

-0.82

LookAnd

-0.81

ècie

-0.80

PhysRevD

-0.77

AndEndTag

-0.76

 Roskov

-0.75

تقاوى

-0.74

POSITIVE LOGITS

for

0.61

to

0.56

 bagi

0.55

 under

0.51

 absolue

0.50

in

0.49

 sekali

0.49

 لنا

0.48

 inmediata

0.48

 assolu

0.47

Activations Density 0.528%

terms that evaluate the practicality and appropriateness of ideas or proposals

possibility or suitability

suitability and value

No Comments

No Known Activations