INDEX

Explanations

mentions of arrests and legal charges

oai_token-act-pair · gpt-3.5-turbo

instances of arrests and related legal terminology

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

natureconservancy

-0.92

Tokens

-0.80

 adapt

-0.75

fried

-0.75

ankind

-0.74

Opt

-0.73

 Remastered

-0.71

 absorbs

-0.71

Reviewed

-0.71

rats

-0.70

POSITIVE LOGITS

 arra

1.15

DUI

1.07

 suspicion

1.06

 felony

1.02

DEA

1.00

okane

1.00

 extradition

0.98

 NYPD

0.94

 Warrant

0.92

 criminally

0.91

Activations Density 0.277%

mentions of arrests and legal charges

instances of arrests and related legal terminology

No Comments

No Known Activations