INDEX

Explanations

the word "oo" with varying levels of activation

oai_token-act-pair · gpt-3.5-turbo

variations of the sound "oo"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

ä»£

-0.70

mus

-0.64

itates

-0.63

ewski

-0.63

 Expend

-0.63

idates

-0.61

 Luthor

-0.60

 oblig

-0.59

adr

-0.59

izoph

-0.58

POSITIVE LOGITS

gey

1.25

gee

1.21

zing

1.21

zie

1.19

ey

1.16

ze

1.16

zy

1.11

zers

1.06

za

1.05

zer

1.05

Activations Density 0.067%

the word "oo" with varying levels of activation

variations of the sound "oo"

No Comments

No Known Activations