INDEX

Explanations

phrases related to hijacking or involvement in hijacking incidents

oai_token-act-pair · gpt-3.5-turbo

terms related to hijacking activities

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Rite

-0.80

 Solitaire

-0.68

ç¥ŀ

-0.66

 Angels

-0.64

-0.63

 Staples

-0.60

 Vlad

-0.60

âĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢâĶĢ

-0.60

 Thick

-0.60

DOC

-0.59

POSITIVE LOGITS

hij

1.12

ackers

1.02

acking

0.93

inx

0.92

acker

0.91

intosh

0.89

adish

0.82

 hijacked

0.82

abis

0.81

aline

0.81

Activations Density 0.020%

phrases related to hijacking or involvement in hijacking incidents

terms related to hijacking activities

No Comments

No Known Activations