INDEX

Explanations

mentions of individuals who have experienced harm or injustice

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "victim" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

leaf

-0.74

ove

-0.71

OPLE

-0.67

REE

-0.65

yip

-0.64

CLASSIFIED

-0.61

 herbal

-0.61

 gust

-0.60

 unconventional

-0.59

 herbs

-0.59

POSITIVE LOGITS

izers

0.92

izes

0.92

vict

0.91

izer

0.89

 victims

0.86

 Victim

0.86

Vict

0.86

 victim

0.86

ization

0.82

 inflicted

0.81

Activations Density 0.047%

mentions of individuals who have experienced harm or injustice

mentions of the word "victim" in various contexts

No Comments

No Known Activations