INDEX

Explanations

phrases and expressions that indicate qualifications or definitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

 ویکی‌پدی

-0.81

 للاسماء

-0.79

transQ

-0.69

UnsafeEnabled

-0.65

 kaarangay

-0.65

 AssemblyCompany

-0.64

tagHelperRunner

-0.63

addContainerGap

-0.63

httphttps

-0.63

 AssemblyTitle

-0.58

POSITIVE LOGITS

 referring

0.37

 gemeint

0.34

Lo

0.33

 means

0.33

ÍAS

0.32

refer

0.31

 meant

0.31

lo

0.31

的意思

0.31

alo

0.30

Activations Density 0.515%

phrases and expressions that indicate qualifications or definitions

No Comments

No Known Activations