INDEX

Explanations

references to returning to a previous state or location

oai_token-act-pair · gpt-3.5-turbo

recurring phrases related to returning to a previous state or place

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

erity

-0.76

Interstitial

-0.71

ander

-0.68

urst

-0.66

illion

-0.65

oS

-0.65

ghai

-0.63

aldi

-0.63

mentation

-0.62

ificant

-0.61

POSITIVE LOGITS

 basics

1.01

 fold

1.00

 original

0.92

 fray

0.87

 drawing

0.86

 haunt

0.83

 roots

0.83

 whence

0.83

 senses

0.81

 birthplace

0.80

Activations Density 0.178%

references to returning to a previous state or location

recurring phrases related to returning to a previous state or place

No Comments

No Known Activations