INDEX

Explanations

keywords related to various diseases

oai_token-act-pair · gpt-3.5-turbo

mentions of diseases

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

oir

-0.70

ynthesis

-0.68

ilts

-0.66

arta

-0.66

tes

-0.66

agall

-0.64

ruce

-0.63

arel

-0.62

itudes

-0.62

endas

-0.61

POSITIVE LOGITS

 disease

1.15

 Disease

0.99

 diseases

0.98

 outbreak

0.98

 infection

0.93

 Outbreak

0.92

 Diseases

0.89

 infections

0.89

 virus

0.88

 illness

0.86

Activations Density 0.011%

keywords related to various diseases

mentions of diseases

No Comments

No Known Activations