INDEX

Explanations

mentions related to physical objects used as props or mentions the term "prop" in various contexts

oai_token-act-pair · gpt-3.5-turbo

references to props and their uses in various contexts

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

 Carbuncle

-0.88

soever

-0.73

Interstitial

-0.71

 Intent

-0.68

 Antar

-0.66

 MacArthur

-0.66

lihood

-0.65

clair

-0.64

 Leaf

-0.64

Maw

-0.64

POSITIVE LOGITS

ulsive

1.25

itious

1.16

etry

1.01

uls

0.95

 prop

0.95

ounded

0.91

hetical

0.89

ulsion

0.84

ension

0.84

asus

0.83

Activations Density 0.010%

mentions related to physical objects used as props or mentions the term "prop" in various contexts

references to props and their uses in various contexts

No Comments

No Known Activations