INDEX

Explanations

words or phrases related to likes, such as "liked" or "likes"

oai_token-act-pair · gpt-3.5-turbo

the presence of segment markers in the text

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĥ¼ãĥĨãĤ£

-0.70

Rim

-0.68

Lit

-0.65

 spirited

-0.63

 Satanic

-0.61

 Labor

-0.61

 compensated

-0.60

éĹĺ

-0.58

ħĭ

-0.58

 Debor

-0.57

POSITIVE LOGITS

ayers

1.35

ocated

1.30

ateral

1.30

oyal

1.28

aptop

1.27

ifestyle

1.27

ibrarian

1.25

ongevity

1.24

aying

1.22

apsed

1.21

Activations Density 0.034%

words or phrases related to likes, such as "liked" or "likes"

the presence of segment markers in the text

No Comments

No Known Activations