INDEX

Explanations

dates and events mentioned in news articles

oai_token-act-pair · gpt-3.5-turbo

phrases and words related to official statements and reporting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

soType

-0.69

cember

-0.57

ļéĨĴ

-0.57

milo

-0.56

Þ

-0.55

StreamerBot

-0.55

ÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤÃĥÃĤ

-0.54

REDACTED

-0.54

 TheNitrome

-0.54

Ą

-0.54

POSITIVE LOGITS

 strokes

0.47

 brittle

0.46

 confess

0.45

 centrif

0.45

 brushes

0.44

 heed

0.44

 scarcely

0.44

 rhythms

0.44

 reckoning

0.43

 plunge

0.43

Activations Density 2.144%

dates and events mentioned in news articles

phrases and words related to official statements and reporting

No Comments

No Known Activations