INDEX

Explanations

terms related to conflict, especially those related to disagreement or opposition

oai_token-act-pair · gpt-3.5-turbo

words related to the concept of "inclusion."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

buster

-0.83

hyde

-0.71

journal

-0.67

 Unch

-0.66

 Hunters

-0.64

culosis

-0.61

 Escape

-0.60

 Masquerade

-0.59

ĸļ

-0.59

 hide

-0.59

POSITIVE LOGITS

idences

1.32

ision

1.29

ursions

1.29

ursion

1.29

itement

1.27

ipient

1.24

identally

1.23

isions

1.21

isive

1.21

orporated

1.18

Activations Density 0.011%

terms related to conflict, especially those related to disagreement or opposition

words related to the concept of "inclusion."

No Comments

No Known Activations