INDEX

Explanations

words related to short-lived phenomena

oai_token-act-pair · gpt-3.5-turbo

terms associated with the concept of being short-lived

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

artisan

-0.72

Union

-0.71

cedented

-0.71

otomy

-0.64

arest

-0.63

condition

-0.60

Ĥª

-0.60

dimension

-0.60

cair

-0.60

HI

-0.59

POSITIVE LOGITS

lived

1.29

uate

0.93

ãĥĥãĤ¯

0.90

ãĤ¤ãĥĪ

0.87

tenance

0.87

lihood

0.86

 Rebell

0.81

uates

0.79

uated

0.77

nesses

0.73

Activations Density 0.011%

words related to short-lived phenomena

terms associated with the concept of being short-lived

No Comments

No Known Activations