INDEX

Explanations

mentions of physical states or actions related to loss of consciousness

oai_token-act-pair · gpt-3.5-turbo

terms related to unconsciousness and states of incapacitation

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

ERO

-0.79

hetti

-0.79

istor

-0.77

risome

-0.73

tein

-0.71

ourke

-0.71

ickr

-0.70

idated

-0.69

adal

-0.67

URI

-0.67

POSITIVE LOGITS

ness

1.10

nesses

0.88

ly

0.81

 unconscious

0.69

lings

0.69

liner

0.66

 bystand

0.64

 baggage

0.63

butt

0.63

LY

0.63

Activations Density 0.013%

mentions of physical states or actions related to loss of consciousness

terms related to unconsciousness and states of incapacitation

No Comments

No Known Activations