INDEX

Explanations

phrases related to political controversies

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

DragonMagazine

-0.96

emi

-0.77

igm

-0.76

 ILCS

-0.75

Contents

-0.74

atan

-0.72

enne

-0.72

imester

-0.72

atch

-0.71

meric

-0.71

POSITIVE LOGITS

 criticizing

1.19

 dispar

1.18

 accusing

1.18

 disdain

1.11

 criticize

1.10

 questioning

1.08

 accuse

1.08

 criticized

1.07

 accuses

1.07

 praising

1.07

Activations Density 0.315%

phrases related to political controversies

No Comments

No Known Activations