INDEX

Explanations

phrases related to actions of self-harm or harm towards others

oai_token-act-pair · gpt-3.5-turbo

references to self-harm and suicide

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 6-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.6.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.6.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

taboola

-0.86

soType

-0.77

eret

-0.74

Management

-0.72

Wide

-0.72

å§«

-0.71

issance

-0.71

DragonMagazine

-0.70

UMP

-0.69

FML

-0.68

POSITIVE LOGITS

 unborn

0.90

 whales

0.75

 innocent

0.74

 hated

0.74

 terrorists

0.73

 classmate

0.73

Mum

0.71

 chickens

0.71

 birds

0.70

 zombies

0.70

Activations Density 0.183%

phrases related to actions of self-harm or harm towards others

references to self-harm and suicide

No Comments

No Known Activations