INDEX

Explanations

words related to negative events or conditions

oai_token-act-pair · gpt-3.5-turbo

references to negative or harmful characteristics and conditions

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

 Carbuncle

-0.81

æĸ¹

-0.73

ALK

-0.72

ORY

-0.70

 Annotations

-0.69

BOOK

-0.67

 FACE

-0.66

 Authorization

-0.66

 Polo

-0.65

 Defenders

-0.65

POSITIVE LOGITS

colm

1.17

ignant

1.15

adies

1.14

practice

1.03

formed

1.02

igned

0.98

arial

0.97

icious

0.96

ady

0.94

absor

0.91

Activations Density 0.011%

words related to negative events or conditions

references to negative or harmful characteristics and conditions

No Comments

No Known Activations