INDEX

Explanations

specific words relating to torture or torment

oai_token-act-pair · gpt-3.5-turbo

terms related to suffering or distress

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

 Consent

-0.69

 Objective

-0.65

 Dakota

-0.64

 recognition

-0.64

PRES

-0.64

OPA

-0.63

soDeliveryDate

-0.62

Suc

-0.62

éļ

-0.61

DragonMagazine

-0.61

POSITIVE LOGITS

onto

1.00

mented

0.91

vell

0.86

seys

0.83

urous

0.82

onite

0.82

ques

0.80

rid

0.79

ritch

0.79

pid

0.78

Activations Density 0.008%

specific words relating to torture or torment

terms related to suffering or distress

No Comments

No Known Activations