INDEX

Explanations

the word "ad" with a strong activation value

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "ad"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

Ń·

-0.85

ĨĴ

-0.67

 sidx

-0.66

 Lovecraft

-0.66

 backwards

-0.65

Ļ½

-0.63

 shorth

-0.62

ĸļ

-0.62

ãĤ¡

-0.62

 Ragnarok

-0.61

POSITIVE LOGITS

venture

1.20

vance

1.17

ventures

1.09

illac

1.05

rill

1.03

apter

1.01

visor

1.01

idas

1.00

rian

1.00

roit

1.00

Activations Density 0.024%

the word "ad" with a strong activation value

occurrences of the word "ad"

No Comments

No Known Activations