INDEX

Explanations

references to software drivers or physical drivers (people driving vehicles) and license-related terms

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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 Legion

-0.73

ertain

-0.72

ritic

-0.72

Lum

-0.72

iversal

-0.70

aeper

-0.70

achu

-0.69

 Seym

-0.67

ormon

-0.67

rients

-0.66

POSITIVE LOGITS

less

0.88

driver

0.84

driving

0.83

 drivers

0.82

 driver

0.81

 license

0.78

 licence

0.78

 Drivers

0.78

 headlights

0.77

License

0.74

Activations Density 0.045%

references to software drivers or physical drivers (people driving vehicles) and license-related terms

references to drivers and driving-related terminology

No Comments

No Known Activations