INDEX

Explanations

references to legal documents or information

oai_token-act-pair · gpt-3.5-turbo

references to ratings or reviews related to movies or productions

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

 Clover

-0.75

glers

-0.74

 Lima

-0.73

creen

-0.68

 Wond

-0.67

Pom

-0.67

renheit

-0.66

 Totem

-0.65

 Granger

-0.64

 Dolphin

-0.64

POSITIVE LOGITS

senal

1.38

ansom

1.09

acing

1.07

agnar

1.05

ICH

1.00

ACK

1.00

ACE

0.98

aced

0.97

ADE

0.97

ascal

0.97

Activations Density 0.024%

references to legal documents or information

references to ratings or reviews related to movies or productions

No Comments

No Known Activations