INDEX

Explanations

mentions of bans or prohibitions

oai_token-act-pair · gpt-3.5-turbo

references to bans or prohibitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 IMAGES

-0.75

 Generations

-0.74

 Apostles

-0.66

 Rhythm

-0.63

 Remastered

-0.61

 rendition

-0.61

 Directors

-0.60

lycer

-0.59

Io

-0.59

 Fired

-0.59

POSITIVE LOGITS

ishment

1.33

hammer

1.12

ishing

1.05

hee

1.00

tering

1.00

zai

1.00

ished

0.93

ish

0.90

nered

0.89

quet

0.87

Activations Density 0.040%

mentions of bans or prohibitions

references to bans or prohibitions

No Comments

No Known Activations