INDEX

Explanations

phrases related to law enforcement activities such as arrests

oai_token-act-pair · gpt-3.5-turbo

occurrences of arrests in the text

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

casters

-0.78

gaard

-0.73

mology

-0.70

vironment

-0.68

issance

-0.67

 Remastered

-0.67

ankind

-0.66

fax

-0.66

emonic

-0.65

pace

-0.65

POSITIVE LOGITS

PKK

0.75

 Constable

0.74

 felon

0.73

 Officers

0.69

 arrested

0.69

 Marshal

0.68

uan

0.66

arnaev

0.65

 criminally

0.65

 protester

0.65

Activations Density 0.050%

phrases related to law enforcement activities such as arrests

occurrences of arrests in the text

No Comments

No Known Activations