INDEX

Explanations

words related to hacking or cyber security

oai_token-act-pair · gpt-3.5-turbo

mentions of the word "Hack" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

iatrics

-0.79

 occas

-0.74

 Archdemon

-0.64

eva

-0.63

userc

-0.62

ittee

-0.62

xus

-0.61

 circulation

-0.61

eous

-0.60

 concess

-0.60

POSITIVE LOGITS

intosh

1.27

athon

1.11

Hack

1.09

aday

0.96

enberg

0.93

ers

0.90

hack

0.86

driver

0.86

lings

0.84

WER

0.83

Activations Density 0.005%

words related to hacking or cyber security

mentions of the word "Hack" in various contexts

No Comments

No Known Activations