INDEX

Explanations

names related to the term "rett", with varying activation values

oai_token-act-pair · gpt-3.5-turbo

mentions of specific names or entities related to events or organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ILCS

-0.77

izoph

-0.75

puter

-0.73

hook

-0.69

ãĤ©

-0.67

undo

-0.66

 foam

-0.65

 Nest

-0.64

bean

-0.64

aroo

-0.63

POSITIVE LOGITS

 Sins

0.78

CLA

0.76

enaries

0.71

 Prosecut

0.70

uary

0.70

sei

0.69

culosis

0.66

 Working

0.66

alez

0.65

agar

0.64

Activations Density 0.053%

names related to the term "rett", with varying activation values

mentions of specific names or entities related to events or organizations

No Comments

No Known Activations