INDEX

Explanations

words related to empowerment and empowering actions

oai_token-act-pair · gpt-3.5-turbo

concepts and discussions related to empowerment

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

patch

-0.83

 Canaver

-0.77

×Ĳ

-0.75

den

-0.72

hiba

-0.70

Son

-0.66

NEY

-0.66

Goo

-0.64

chal

-0.64

hound

-0.63

POSITIVE LOGITS

ments

1.00

Reviewer

0.83

ment

0.82

ittees

0.76

mentation

0.75

 empower

0.74

MENTS

0.73

EStream

0.73

 empowered

0.72

iences

0.72

Activations Density 0.013%

words related to empowerment and empowering actions

concepts and discussions related to empowerment

No Comments

No Known Activations