INDEX

Explanations

autoantibodies and terms related to autoimmunity

oai_token-act-pair · gpt-3.5-turbo

terms related to automobiles and auto-related topics

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

ĸļ

-1.22

soever

-0.92

enance

-0.92

ividual

-0.89

 GOODMAN

-0.87

edin

-0.84

nesday

-0.84

hips

-0.82

lihood

-0.80

swick

-0.80

POSITIVE LOGITS

immune

0.95

cephal

0.87

 autos

0.76

 bailout

0.74

 accident

0.73

actively

0.72

exec

0.70

incre

0.70

 appra

0.69

 accidents

0.68

Activations Density 0.013%

autoantibodies and terms related to autoimmunity

terms related to automobiles and auto-related topics

No Comments

No Known Activations