INDEX

Explanations

verbs related to causing intense emotional reactions

oai_token-act-pair · gpt-3.5-turbo

instances of the word "implicate" or its variations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

chal

-0.79

Ĥ¬

-0.75

kees

-0.74

grass

-0.72

liness

-0.71

lette

-0.65

flix

-0.65

 cleaner

-0.64

grab

-0.63

 Warwick

-0.63

POSITIVE LOGITS

osion

1.54

oded

1.47

ausible

1.43

icating

1.34

ications

1.28

icate

1.27

icates

1.27

oding

1.27

icit

1.25

anting

1.18

Activations Density 0.025%

verbs related to causing intense emotional reactions

instances of the word "implicate" or its variations

No Comments

No Known Activations