INDEX

Explanations

information related to processed foods or substances

oai_token-act-pair · gpt-3.5-turbo

references to processed foods

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

liest

-0.73

lust

-0.65

fall

-0.65

ansky

-0.65

gets

-0.63

 Lazarus

-0.62

show

-0.61

drops

-0.60

bie

-0.60

bors

-0.59

POSITIVE LOGITS

asonic

0.86

 meats

0.84

icer

0.82

avascript

0.81

 Processing

0.80

chnology

0.79

atoes

0.78

 processing

0.78

ascript

0.78

 processed

0.78

Activations Density 0.018%

information related to processed foods or substances

references to processed foods

No Comments

No Known Activations