INDEX

Explanations

words related to bigotry, discrimination, and negative attitudes towards certain groups or individuals

oai_token-act-pair · gpt-3.5-turbo

references to bigotry and related negative behaviors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 confir

-0.71

pring

-0.69

 Cancel

-0.69

 Accounting

-0.67

ournal

-0.67

OA

-0.66

birth

-0.66

 synchronization

-0.63

 Contracts

-0.63

%"

-0.63

POSITIVE LOGITS

uously

0.86

ifiers

0.84

 bigotry

0.82

etooth

0.80

 bigot

0.80

ifying

0.78

ifiable

0.78

itude

0.77

izer

0.75

itious

0.75

Activations Density 0.013%

words related to bigotry, discrimination, and negative attitudes towards certain groups or individuals

references to bigotry and related negative behaviors

No Comments

No Known Activations