INDEX

Explanations

words related to content being removed or taken down

oai_token-act-pair · gpt-3.5-turbo

references to social media interactions and viral 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Experience

-0.72

 accommodations

-0.72

 sacrifices

-0.71

letico

-0.71

 accommodation

-0.68

 Instrument

-0.66

ints

-0.66

 concessions

-0.65

 autonomy

-0.65

 captains

-0.65

POSITIVE LOGITS

 archived

1.55

 published

1.53

 reprinted

1.52

 circulated

1.50

 uploaded

1.46

 posted

1.44

 deleted

1.41

 updated

1.31

 downloaded

1.31

 cached

1.29

Activations Density 0.339%

words related to content being removed or taken down

references to social media interactions and viral content

No Comments

No Known Activations