INDEX

Explanations

phrases indicating the expression of opinions or concerns by individuals or groups

oai_token-act-pair · gpt-3.5-turbo

instances of the word "expressed" and related forms

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

é¾įå

-0.77

ade

-0.73

ctors

-0.72

gey

-0.69

osate

-0.68

fman

-0.65

capt

-0.63

ADE

-0.63

»Ĵ

-0.63

strap

-0.63

POSITIVE LOGITS

 displeasure

1.50

 frustration

1.33

 gratitude

1.33

 concerns

1.30

 concern

1.25

 willingness

1.25

 frustrations

1.25

 doubts

1.24

 remorse

1.22

 admiration

1.22

Activations Density 0.099%

phrases indicating the expression of opinions or concerns by individuals or groups

instances of the word "expressed" and related forms

No Comments

No Known Activations