INDEX

Explanations

words related to actions of suppression or restriction

oai_token-act-pair · gpt-3.5-turbo

words or phrases related to questioning or uncertainty

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

 Prosper

-0.66

strap

-0.65

 hist

-0.65

 permitting

-0.64

 stun

-0.62

 breeze

-0.61

 disappoint

-0.61

roy

-0.61

 overhead

-0.60

 TODAY

-0.60

POSITIVE LOGITS

ibble

1.32

irk

1.30

arre

1.30

ivering

1.27

iets

1.18

arant

1.17

orum

1.17

itter

1.17

ilt

1.15

ilts

1.12

Activations Density 0.015%

words related to actions of suppression or restriction

words or phrases related to questioning or uncertainty

No Comments

No Known Activations