INDEX

Explanations

phrases related to legal or law enforcement activities

oai_token-act-pair · gpt-3.5-turbo

mentions of "authorities."

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

thal

-0.74

eem

-0.74

ces

-0.69

 Candle

-0.69

leaf

-0.66

tom

-0.64

Bow

-0.64

rous

-0.64

 Trem

-0.63

TON

-0.63

POSITIVE LOGITS

 authorities

1.11

 agencies

0.96

 Authorities

0.95

 confir

0.88

 officials

0.88

 tasked

0.83

 decree

0.80

 resorted

0.78

 consulted

0.78

 Officials

0.78

Activations Density 0.020%

phrases related to legal or law enforcement activities

mentions of "authorities."

No Comments

No Known Activations