INDEX

Explanations

words related to medical conditions, specifically focusing on the liver

oai_token-act-pair · gpt-3.5-turbo

references to liver health and related medical issues

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

minus

-0.80

 Wildcats

-0.72

anamo

-0.71

ãģ®éŃĶ

-0.70

é¾įå¥ĳå£«

-0.70

Ans

-0.70

Christian

-0.70

Fif

-0.70

 Occupations

-0.68

 Subtle

-0.65

POSITIVE LOGITS

 liver

1.08

cir

1.00

 enzymes

0.98

 detox

0.95

 transplant

0.92

 enzyme

0.88

 hepat

0.87

 gland

0.82

 glands

0.81

etics

0.81

Activations Density 0.007%

words related to medical conditions, specifically focusing on the liver

references to liver health and related medical issues

No Comments

No Known Activations