INDEX

Explanations

phrases indicating certainty or affirmation

oai_token-act-pair · gpt-3.5-turbo

the word "Sure" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

foreseen

-0.71

 âĢİ

-0.65

 contrace

-0.63

exclusive

-0.59

 psychiat

-0.58

 recomm

-0.56

":""},{"

-0.56

govtrack

-0.56

ublic

-0.56

POSITIVE LOGITS

 enough

1.25

ndra

1.06

ly

1.03

footed

0.98

enough

0.97

 Enough

0.85

ty

0.84

terday

0.81

fire

0.80

 sounds

0.78

Activations Density 0.019%

phrases indicating certainty or affirmation

the word "Sure" in various contexts

No Comments

No Known Activations