INDEX

Explanations

phrases related to various forms of attacks and attackers

oai_token-act-pair · gpt-3.5-turbo

references to individuals involved in violent actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

zl

-0.97

umph

-0.84

Balt

-0.84

 Revival

-0.76

sit

-0.75

urgical

-0.71

wheel

-0.71

 rebirth

-0.69

urb

-0.69

indust

-0.67

POSITIVE LOGITS

 attacker

0.86

 attackers

0.85

 beware

0.80

 wielding

0.76

 attacked

0.75

 assailants

0.75

 assailant

0.72

 intent

0.72

who

0.71

 rapist

0.71

Activations Density 0.029%

phrases related to various forms of attacks and attackers

references to individuals involved in violent actions or events

No Comments

No Known Activations