INDEX

Explanations

instances where an action is enabled or allowed

oai_token-act-pair · gpt-3.5-turbo

phrases that convey enabling or allowing actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

 Languages

-0.70

Zen

-0.68

bage

-0.68

querque

-0.63

 Beasts

-0.61

lihood

-0.60

liest

-0.60

imposed

-0.59

externalActionCode

-0.59

Øª

-0.58

POSITIVE LOGITS

tered

1.20

tering

1.11

icia

0.87

eries

0.74

 loose

0.74

itia

0.73

isters

0.73

us

0.73

 slip

0.72

enne

0.71

Activations Density 0.024%

instances where an action is enabled or allowed

phrases that convey enabling or allowing actions

No Comments

No Known Activations