INDEX

Explanations

words related to security

oai_token-act-pair · gpt-3.5-turbo

the prefix "sc" indicating words related to "screen" or "scenarios."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

Downloadha

-0.75

EStream

-0.70

 remembrance

-0.70

 preempt

-0.68

 disarm

-0.65

 brisk

-0.64

 gent

-0.64

 Goose

-0.63

 Lauder

-0.63

 reconstruction

-0.62

POSITIVE LOGITS

apers

1.22

reens

1.21

rawl

1.20

ratch

1.19

rupulous

1.17

opes

1.15

urry

1.09

anned

1.07

aped

1.06

rawled

1.06

Activations Density 0.009%

words related to security

the prefix "sc" indicating words related to "screen" or "scenarios."

No Comments

No Known Activations