INDEX

Explanations

expressions or statements of disapproval or condemnation

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "denounce" and its variations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

olen

-0.87

omaly

-0.83

ramid

-0.83

ammy

-0.78

cially

-0.78

aker

-0.76

perm

-0.73

icum

-0.72

OVA

-0.72

thora

-0.71

POSITIVE LOGITS

 denouncing

0.98

 denounce

0.89

 denounced

0.81

 loudly

0.80

 disav

0.74

 aloud

0.73

urous

0.72

 racism

0.71

 condemning

0.69

 homophobia

0.68

Activations Density 0.013%

expressions or statements of disapproval or condemnation

occurrences of the word "denounce" and its variations

No Comments

No Known Activations