INDEX

Explanations

words related to political activities or events

oai_token-act-pair · gpt-3.5-turbo

references to legislation or political agreements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 americ

-0.94

 dudes

-0.88

liv

-0.87

 dude

-0.83

ukemia

-0.82

 someday

-0.82

 traveler

-0.80

 goddamn

-0.79

 marvelous

-0.78

Kids

-0.76

POSITIVE LOGITS

 criticised

1.27

 manoeuv

1.23

 scept

1.14

 organisers

1.07

DUP

1.06

MPs

1.04

 signalling

1.04

 emphas

1.03

 organis

1.03

 Nato

1.02

Activations Density 0.911%

words related to political activities or events

references to legislation or political agreements

No Comments

No Known Activations