INDEX

Explanations

words related to personal attributes or possessions

oai_token-act-pair · gpt-3.5-turbo

terms related to personal information and data

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

xual

-1.18

UMP

-0.78

 âĶľ

-0.71

NESS

-0.71

REG

-0.68

å¾

-0.68

 Twain

-0.68

GGGG

-0.67

LER

-0.67

DERR

-0.67

POSITIVE LOGITS

ised

1.27

ization

1.14

ized

1.12

izing

1.11

ities

1.06

isation

1.05

izations

1.03

izes

1.03

isations

1.02

istic

1.01

Activations Density 0.026%

words related to personal attributes or possessions

terms related to personal information and data

No Comments

No Known Activations