INDEX

Explanations

verbs related to discovering, learning, revealing, creating, and finding

oai_token-act-pair · gpt-3.5-turbo

phrases indicating discovery or revelation of information

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

click

-0.63

 suffice

-0.59

opter

-0.58

<<

-0.55

sis

-0.55

Ã´

-0.55

||

-0.54

craft

-0.54

itas

-0.54

Cre

-0.54

POSITIVE LOGITS

interstitial

0.74

iosyncr

0.68

 anecd

0.67

autions

0.66

ractive

0.65

soDeliveryDate

0.64

 Cosponsors

0.64

saf

0.63

 endorsements

0.63

 Tradable

0.63

Activations Density 0.689%

verbs related to discovering, learning, revealing, creating, and finding

phrases indicating discovery or revelation of information

No Comments

No Known Activations