INDEX

Explanations

instances of the word "knew."

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

knew

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

存于互联网档案馆

-0.57

mathbf

-0.51

fortawesome

-0.47

 Plata

-0.47

 argint

-0.47

 dreapta

-0.46

operatorname

-0.46

 jazdy

-0.43

 stanga

-0.42

olverine

-0.42

POSITIVE LOGITS

 Knew

0.84

knew

0.70

 knew

0.63

 Drew

0.62

grew

0.62

Drew

0.61

uuu

0.59

Kew

0.59

drew

0.58

rew

0.57

Activations Density 0.012%

instances of the word "knew."

knew

No Comments

No Known Activations