INDEX

Explanations

instances where individuals conceal or reveal information due to fear or security concerns

oai_token-act-pair · gpt-3.5-turbo

instances of anonymity and security-related concerns

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

average

-0.64

ibles

-0.60

 erupt

-0.58

orting

-0.58

 explodes

-0.57

opath

-0.57

 engulfed

-0.57

ail

-0.57

iped

-0.57

 dominates

-0.56

POSITIVE LOGITS

 precaution

1.27

 fearing

1.25

because

1.10

 because

1.09

bec

1.07

Because

1.02

 lest

0.97

 fear

0.95

 Because

0.94

ecause

0.91

Activations Density 1.101%

instances where individuals conceal or reveal information due to fear or security concerns

instances of anonymity and security-related concerns

No Comments

No Known Activations