INDEX

Explanations

mentions of "whole" in relation to foods or health

oai_token-act-pair · gpt-3.5-turbo

references to "whole" foods or whole ingredients

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

yip

-0.72

eer

-0.66

ministic

-0.63

 Superintendent

-0.63

ptive

-0.62

geon

-0.60

 Prel

-0.60

 Examiner

-0.59

KER

-0.58

endants

-0.57

POSITIVE LOGITS

heartedly

1.70

hearted

1.21

meal

0.95

allo

0.79

ãĤ¨ãĥ«

0.77

OME

0.75

oton

0.74

ciating

0.73

whe

0.71

 Foods

0.70

Activations Density 0.018%

mentions of "whole" in relation to foods or health

references to "whole" foods or whole ingredients

No Comments

No Known Activations