INDEX

Explanations

the word "specter" or variations of it

oai_token-act-pair · gpt-3.5-turbo

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

side

-0.64

etheless

-0.63

mint

-0.63

NCT

-0.60

backer

-0.60

heads

-0.59

 LIFE

-0.58

 Springer

-0.56

nesses

-0.56

capacity

-0.54

POSITIVE LOGITS

acular

1.36

rums

1.19

acle

1.13

acles

1.12

rosc

1.06

rum

1.04

rogram

1.03

ators

0.97

uous

0.96

ral

0.92

Activations Density 0.048%

the word "specter" or variations of it

No Comments

No Known Activations