INDEX

Explanations

addresses, locations, or directions involving "Drive"

oai_token-act-pair · gpt-3.5-turbo

references to "Drive," possibly indicating locations or directions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

 Seym

-1.05

icipated

-0.85

ileaks

-0.78

iannopoulos

-0.74

krit

-0.72

ANN

-0.70

wealth

-0.69

 constitu

-0.68

ambo

-0.67

abad

-0.66

POSITIVE LOGITS

train

0.87

away

0.82

bys

0.78

 drives

0.77

Drive

0.75

wheel

0.73

bike

0.72

 thru

0.71

driving

0.71

 drive

0.70

Activations Density 0.023%

addresses, locations, or directions involving "Drive"

references to "Drive," possibly indicating locations or directions

No Comments

No Known Activations