INDEX

Explanations

information related to drivers or driving

oai_token-act-pair · gpt-3.5-turbo

references to drivers and driving-related terminology

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

 Flavoring

-0.82

aeper

-0.80

 Seym

-0.79

iversal

-0.79

ertain

-0.77

ritic

-0.76

yss

-0.76

ropolitan

-0.72

achu

-0.72

amorph

-0.68

POSITIVE LOGITS

 drivers

0.90

 pige

0.87

 driver

0.86

less

0.82

driver

0.82

driving

0.80

boats

0.78

 beware

0.76

 whales

0.76

 headlights

0.76

Activations Density 0.022%

information related to drivers or driving

references to drivers and driving-related terminology

No Comments

No Known Activations