INDEX

Explanations

vocabulary related to censorship or filtering

oai_token-act-pair · gpt-3.5-turbo

terms related to censorship and its implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 Scotia

-0.64

ordinary

-0.63

amaz

-0.60

holding

-0.59

sheet

-0.59

zzi

-0.58

 Tinker

-0.58

addafi

-0.57

agne

-0.56

 McAuliffe

-0.55

POSITIVE LOGITS

chen

0.98

orship

0.97

ource

0.96

manship

0.94

hift

0.92

terday

0.91

urable

0.90

wear

0.88

CRIP

0.85

haw

0.85

Activations Density 0.050%

vocabulary related to censorship or filtering

terms related to censorship and its implications

No Comments

No Known Activations