INDEX

Explanations

terms relating to safe, private, or desirable locations

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70

Prompts (Dashboard)

36,864 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

WEBPACK

-0.77

NewUrlParser

-0.68

 jLabel

-0.62

undra

-0.59

WebMethod

-0.56

ndice

-0.55

 Faced

-0.54

 wobec

-0.54

igshid

-0.54

companying

-0.53

POSITIVE LOGITS

 where

0.73

où

0.61

verwijspagina

0.60

ValueStyle

0.60

 conducive

0.59

GOTREF

0.58

 donde

0.58

where

0.55

 tempat

0.54

Where

0.53

Activations Density 0.161%

terms relating to safe, private, or desirable locations

No Comments

No Known Activations