INDEX

Explanations

instances of the word "wrote" and related writing actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

ParallelGroup

-0.68

 manqué

-0.56

DataAnnotations

-0.54

 hindurch

-0.52

Portail

-0.49

endence

-0.49

Tazama

-0.49

árbol

-0.49

بوابة

-0.49

egy

-0.49

POSITIVE LOGITS

EDEFAULT

0.77

MessageTagHelper

0.73

 CURIAM

0.58

IntoConstraints

0.56

abestanden

0.54

 EconPapers

0.53

 autorytatywna

0.53

bouts

0.53

 snippetHide

0.53

#+#

0.52

Activations Density 0.088%

instances of the word "wrote" and related writing actions

No Comments

No Known Activations