INDEX

Explanations

specific and unusual word forms or variations that express complexity or nuance

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

 OFDb

-0.66

 Monfieur

-0.63

Personendaten

-0.59

OrNil

-0.59

 myſelf

-0.58

 ſeveral

-0.57

ativement

-0.57

istiques

-0.56

Vidite

-0.56

 stethoscope

-0.55

POSITIVE LOGITS

ванович

0.51

archiviato

0.48

ürm

0.48

atorze

0.47

EventLoop

0.47

 censi

0.47

cinct

0.46

ulement

0.46

enerbah

0.45

guenos

0.45

Activations Density 0.418%

specific and unusual word forms or variations that express complexity or nuance

No Comments

No Known Activations