INDEX

Explanations

words related to dogs and their characteristics or behaviors

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

rativa

-0.52

{}{}

-0.48

 Mercier

-0.47

lccccc

-0.47

 autorytatywna

-0.47

 amass

-0.47

AMI

-0.46

 invokingState

-0.46

Kei

-0.45

++];

-0.43

POSITIVE LOGITS

 Dogs

1.33

Dog

1.30

Dog

1.30

 dogs

1.26

dog

1.25

DOG

1.17

Dogs

1.16

dog

1.08

 DOGS

1.05

DOG

1.05

Activations Density 0.012%

words related to dogs and their characteristics or behaviors

No Comments

No Known Activations