INDEX

Explanations

words related to repealing or revealing information

oai_token-act-pair · gpt-3.5-turbo

words related to the concept of repeal and revealing information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

istically

-0.74

icum

-0.72

istic

-0.67

 stature

-0.67

istics

-0.65

 foul

-0.64

iates

-0.64

Io

-0.61

 Danish

-0.61

é¾įå

-0.59

POSITIVE LOGITS

lling

1.78

llers

1.66

ller

1.57

lled

1.51

xual

1.31

aled

1.25

aling

1.21

rent

1.19

ll

1.18

aters

1.11

Activations Density 0.061%

words related to repealing or revealing information

words related to the concept of repeal and revealing information

No Comments

No Known Activations