INDEX

Explanations

phrases related to advertisements and possibly company names

oai_token-act-pair · gpt-3.5-turbo

occurrences of advertisements and marketing-related content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 tram

-0.88

 hospitality

-0.87

 advoc

-0.87

 princ

-0.86

 citiz

-0.85

 mathemat

-0.83

 enriched

-0.83

 mutually

-0.81

 petrol

-0.80

 partner

-0.80

POSITIVE LOGITS

Advertisement

2.08

RELATED

1.55

advertisement

1.50

MORE

1.38

According

1.37

Alert

1.35

Anyway

1.35

Photo

1.34

Update

1.33

That

1.32

Activations Density 0.066%

phrases related to advertisements and possibly company names

occurrences of advertisements and marketing-related content

No Comments

No Known Activations