INDEX

Explanations

references to dead bodies or corpses

oai_token-act-pair · gpt-3.5-turbo

references to corpses and bodies in graphic or macabre contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

orney

-0.84

itsch

-0.76

Americ

-0.72

 Monthly

-0.72

itudes

-0.69

 Berks

-0.68

rish

-0.65

ipolar

-0.64

Channel

-0.64

wana

-0.63

POSITIVE LOGITS

 corpses

1.13

 carc

1.09

 corpse

1.07

 Corpse

1.00

 slain

0.82

 rotting

0.81

 pile

0.80

 Reaper

0.77

 skulls

0.77

lly

0.76

Activations Density 0.034%

references to dead bodies or corpses

references to corpses and bodies in graphic or macabre contexts

No Comments

No Known Activations