INDEX

Explanations

mentions of drivers in various contexts

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "driver."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 Flavoring

-0.77

oho

-0.76

ropolitan

-0.74

 Seym

-0.74

orp

-0.72

achu

-0.70

oultry

-0.69

anamo

-0.68

akings

-0.67

ablishment

-0.66

POSITIVE LOGITS

 driver

1.30

driver

1.29

 drivers

1.17

Driver

1.00

drivers

0.95

 Drivers

0.94

 Driver

0.92

lane

0.88

pool

0.82

driving

0.82

Activations Density 0.011%

mentions of drivers in various contexts

occurrences of the word "driver."

No Comments

No Known Activations