INDEX

Explanations

mentions of large entities or corporations

oai_token-act-pair · gpt-3.5-turbo

references to "big" entities, particularly in the context of business and politics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

cia

-0.70

 Reincarnated

-0.69

DragonMagazine

-0.69

 Emin

-0.68

Dialogue

-0.67

 Cheong

-0.66

yrinth

-0.65

veyard

-0.65

 Michaels

-0.65

IRC

-0.64

POSITIVE LOGITS

oted

1.25

wig

1.01

gest

1.01

 bucks

0.94

 chunks

0.91

 chunk

0.91

GER

0.88

ol

0.87

 enough

0.85

ger

0.85

Activations Density 0.061%

mentions of large entities or corporations

references to "big" entities, particularly in the context of business and politics

No Comments

No Known Activations