INDEX

Explanations

legal terms or actions related to law enforcement, such as "warrant."

oai_token-act-pair · gpt-3.5-turbo

mentions of warrants in relation to law enforcement activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

hedral

-0.72

thumbnails

-0.72

 Tycoon

-0.67

ttle

-0.66

 Chick

-0.63

 Dunk

-0.62

 Reaction

-0.61

helle

-0.61

 Norn

-0.61

 Survivors

-0.59

POSITIVE LOGITS

 warrant

1.19

 warrants

1.08

lessly

0.92

 affidavit

0.87

 authorizing

0.86

 Warrant

0.82

 forwarding

0.81

 authorization

0.80

 affidav

0.79

ably

0.78

Activations Density 0.016%

legal terms or actions related to law enforcement, such as "warrant."

mentions of warrants in relation to law enforcement activities

No Comments

No Known Activations