INDEX

Explanations

words related to attackers and attacks

oai_token-act-pair · gpt-3.5-turbo

references to an "attacker."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

zl

-0.98

wheel

-0.78

REE

-0.73

umph

-0.72

sit

-0.70

Balt

-0.70

 unification

-0.70

spring

-0.69

carb

-0.68

 Revival

-0.67

POSITIVE LOGITS

 attacker

1.00

 attackers

0.98

 attacked

0.85

 assailant

0.79

 assailants

0.79

 attack

0.79

 penetrated

0.74

CVE

0.71

 perpetrated

0.71

 beware

0.71

Activations Density 0.015%

words related to attackers and attacks

references to an "attacker."

No Comments

No Known Activations