INDEX

Explanations

phrases related to giving or receiving instructions, commands, or feedback containing verbs

oai_token-act-pair · gpt-3.5-turbo

expressions of confrontation or challenge

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

Eastern

-0.71

 Coliseum

-0.70

Ĥª

-0.66

uyomi

-0.65

OSP

-0.64

Patch

-0.63

burgh

-0.61

thodox

-0.61

ACP

-0.60

lvl

-0.60

POSITIVE LOGITS

 yourselves

1.71

 yourself

1.30

 your

0.88

 Yourself

0.86

sir

0.80

 YOUR

0.80

 retard

0.78

your

0.76

 cens

0.72

 majesty

0.70

Activations Density 0.357%

phrases related to giving or receiving instructions, commands, or feedback containing verbs

expressions of confrontation or challenge

No Comments

No Known Activations