INDEX

Explanations

words related to insults or derogatory remarks

oai_token-act-pair · gpt-3.5-turbo

references to insults or derogatory language

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

arijuana

-0.82

-0.78

ilver

-0.75

negie

-0.74

etheus

-0.74

aver

-0.72

ccording

-0.70

angler

-0.68

 Folder

-0.67

agically

-0.66

POSITIVE LOGITS

 insult

1.37

 insults

1.17

 insulted

1.16

 insulting

1.05

ingly

0.97

 disrespect

0.97

 humili

0.92

 prejudice

0.89

 offend

0.86

ãĤ¹ãĥĪ

0.86

Activations Density 0.015%

words related to insults or derogatory remarks

references to insults or derogatory language

No Comments

No Known Activations