INDEX

Explanations

phrases related to being punished or reprimanded

oai_token-act-pair · gpt-3.5-turbo

references to slapping or related actions and their implications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

tis

-0.76

icult

-0.75

ernand

-0.71

ests

-0.68

EMBER

-0.67

oÄŁ

-0.61

éĸ

-0.61

 Seventh

-0.59

mpeg

-0.59

 Grande

-0.59

POSITIVE LOGITS

dash

1.20

dab

1.07

creen

1.05

stick

0.94

down

0.75

lihood

0.75

brush

0.73

bang

0.73

ction

0.73

metry

0.72

Activations Density 0.077%

phrases related to being punished or reprimanded

references to slapping or related actions and their implications

No Comments

No Known Activations