INDEX

Explanations

phrases related to personal aspects or attributes

oai_token-act-pair · gpt-3.5-turbo

references to personal attributes or experiences

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

xual

-1.13

ĸļ

-0.77

UMP

-0.75

å¾

-0.72

 Archdemon

-0.72

REG

-0.71

LER

-0.71

ÄŁ

-0.70

YP

-0.70

UP

-0.69

POSITIVE LOGITS

ised

1.32

ities

1.16

ization

1.15

ized

1.13

isation

1.12

izations

1.09

 belongings

1.07

isations

1.05

izing

1.03

 hygiene

1.02

Activations Density 0.038%

phrases related to personal aspects or attributes

references to personal attributes or experiences

No Comments

No Known Activations