INDEX

Explanations

phrases related to disciplinary actions

oai_token-act-pair · gpt-3.5-turbo

terms related to disciplinary actions

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

sonian

-0.79

Phones

-0.74

htaking

-0.71

stanbul

-0.69

hing

-0.69

DonaldTrump

-0.69

ï¸ı

-0.67

imble

-0.66

frey

-0.66

egg

-0.66

POSITIVE LOGITS

 disciplinary

1.18

 proceedings

0.96

 sanction

0.89

 punishments

0.88

 inquiry

0.87

 discipline

0.85

 hearings

0.85

 punishment

0.84

 inquiries

0.83

ities

0.82

Activations Density 0.014%

phrases related to disciplinary actions

terms related to disciplinary actions

No Comments

No Known Activations