INDEX

Explanations

words related to health conditions, especially strokes

oai_token-act-pair · gpt-3.5-turbo

references to "strokes" in various contexts, particularly medical conditions and actions

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

guyen

-0.79

ãĤ©

-0.74

%"

-0.72

gemony

-0.70

etheless

-0.69

essor

-0.63

ership

-0.63

merce

-0.62

ilogy

-0.61

itive

-0.61

POSITIVE LOGITS

 stroke

1.10

 strokes

0.97

manship

0.89

stroke

0.81

 Stro

0.76

aukee

0.74

weed

0.73

teenth

0.70

utic

0.69

 Prevention

0.67

Activations Density 0.020%

words related to health conditions, especially strokes

references to "strokes" in various contexts, particularly medical conditions and actions

No Comments

No Known Activations