INDEX

Explanations

general quantifiers indicating the presence of items or concepts

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

"In all" or "in some" followed by a noun

np_acts-logits-general · gemini-2.0-flash

in all cases

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

 purpoſe

-0.59

enterOuterAlt

-0.59

 مشين

-0.58

 houſe

-0.57

 themſelves

-0.57

 ſch

-0.56

TSR

-0.55

 itſelf

-0.55

ChildScrollView

-0.55

POSITIVE LOGITS

 cases

0.76

 parts

0.70

 phases

0.67

 Italijanski

0.59

 stages

0.59

 instances

0.59

ILLER

0.56

 places

0.55

AKES

0.54

ueba

0.54

Activations Density 0.199%

general quantifiers indicating the presence of items or concepts

"In all" or "in some" followed by a noun

in all cases

No Comments

No Known Activations