INDEX

Explanations

scientific terms and research findings related to medical or psychological studies

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate research findings or correlations in studies

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

udence

-0.84

ð

-0.77

rehens

-0.76

wered

-0.75

erity

-0.72

raphics

-0.71

 IMAGES

-0.69

numbered

-0.68

sembly

-0.68

Hard

-0.67

POSITIVE LOGITS

 causal

1.60

 causation

1.42

 correlation

1.41

 caus

1.40

 correlations

1.31

 mechanisms

1.30

 correlates

1.23

 synerg

1.23

 biomark

1.23

 link

1.22

Activations Density 0.265%

scientific terms and research findings related to medical or psychological studies

phrases that indicate research findings or correlations in studies

No Comments

No Known Activations