INDEX

Explanations

terms related to legal actions and consequences, specifically focusing on words like "recourse," "impunity," and "prosecute."

oai_token-act-pair · gpt-3.5-turbo

terms related to legal actions and consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

itect

-0.84

 Fantastic

-0.71

depend

-0.68

ravis

-0.66

 dehyd

-0.65

 Zone

-0.65

 Optim

-0.65

 Delta

-0.65

 Panda

-0.65

 Stranger

-0.64

POSITIVE LOGITS

 recourse

2.62

 impunity

1.80

 prosecute

1.63

 prosecuted

1.52

 prosec

1.30

 avenues

1.28

 prosecutions

1.26

 avenue

1.23

 prosecuting

1.21

 redress

1.18

Activations Density 0.058%

terms related to legal actions and consequences, specifically focusing on words like "recourse," "impunity," and "prosecute."

terms related to legal actions and consequences

No Comments

No Known Activations