INDEX

Explanations

activations related to physical movement and actions involving people

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 23-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_23/width_16k/average_l0_75

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.23.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

Personensuche

-0.90

rrggbb

-0.55

!*\

-0.53

tvguidetime

-0.52

 setuptools

-0.51

WriteBarrier

-0.51

ReusableCell

-0.50

 onAnimation

-0.50

Portály

-0.49

DataAnnotations

-0.49

POSITIVE LOGITS

 downstairs

0.70

 upstairs

0.64

 running

0.64

 corriendo

0.60

 RUNNING

0.58

 rushing

0.56

 walking

0.55

 Running

0.55

 hurrying

0.54

+#+

0.53

Activations Density 0.138%

activations related to physical movement and actions involving people

No Comments

No Known Activations