INDEX

Explanations

"St" followed by an English word, possibly in the context of articles, studies, or reports

oai_token-act-pair · gpt-3.5-turbo

the end of textual content or formatting markers

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

ļéĨĴ

-0.72

hower

-0.71

hound

-0.64

gee

-0.64

EStream

-0.64

 thumbs

-0.60

 friendly

-0.57

ciation

-0.57

cules

-0.57

 deaf

-0.56

POSITIVE LOGITS

upid

1.34

rict

1.28

uffed

1.24

unning

1.22

uart

1.19

ructure

1.14

arters

1.14

alker

1.11

rikes

1.10

adium

1.08

Activations Density 0.033%

"St" followed by an English word, possibly in the context of articles, studies, or reports

the end of textual content or formatting markers

No Comments

No Known Activations