INDEX

Explanations

references to sources and citations

oai_token-act-pair · gpt-3.5-turbo

connections and relationships between phrases and categories in informational content

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

\\\\\\\\

-0.68

ONSORED

-0.63

è»

-0.62

hower

-0.60

iHUD

-0.58

ãĤ¼ãĤ¦ãĤ¹

-0.58

Ô

-0.56

soDeliveryDate

-0.55

issance

-0.55

Leaks

-0.54

POSITIVE LOGITS

ettings

0.59

ynchron

0.57

 prolifer

0.56

 respectively

0.56

 backgrounds

0.55

 converge

0.53

hett

0.53

 stray

0.53

 arsen

0.52

ongyang

0.51

Activations Density 2.721%

references to sources and citations

connections and relationships between phrases and categories in informational content

No Comments

No Known Activations