INDEX

Explanations

news or media snippets related to unfortunate events or controversy

oai_token-act-pair · gpt-3.5-turbo

sentences or phrases that indicate the conclusion of an event or statement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 charact

-0.84

 instinct

-0.80

 unch

-0.71

 defe

-0.70

 unden

-0.70

 unequ

-0.68

 minion

-0.66

 deserts

-0.66

 gorge

-0.66

 volunte

-0.65

POSITIVE LOGITS

 Similarly

1.42

 Another

1.32

 Later

1.26

 Similar

1.25

 Likewise

1.23

 Meanwhile

1.22

 Earlier

1.17

 Additionally

1.15

Similarly

1.11

 Previously

1.10

Activations Density 0.491%

news or media snippets related to unfortunate events or controversy

sentences or phrases that indicate the conclusion of an event or statement

No Comments

No Known Activations