INDEX

Explanations

actions related to rules or authority

oai_token-act-pair · gpt-3.5-turbo

actions that involve making mistakes or facing consequences

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ilogy

-0.73

inger

-0.70

DragonMagazine

-0.70

erm

-0.70

Depending

-0.66

ciating

-0.65

endif

-0.65

mort

-0.64

fortunately

-0.64

alg

-0.64

POSITIVE LOGITS

 improper

0.80

 illegal

0.76

BuyableInstoreAndOnline

0.71

ãĤ¤ãĥĪ

0.70

 illegally

0.66

 violations

0.65

 unsafe

0.65

 insufficient

0.64

 excessively

0.63

 improperly

0.62

Activations Density 0.187%

actions related to rules or authority

actions that involve making mistakes or facing consequences

No Comments

No Known Activations