INDEX

Explanations

personal interactions and conversations involving specific actions and emotions

oai_token-act-pair · gpt-3.5-turbo

instances of reported speech or claims made by individuals

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

 rall

-0.70

emetery

-0.67

espie

-0.65

 hindsight

-0.65

ements

-0.65

 Coliseum

-0.65

-+-+

-0.64

SDL

-0.64

 Shogun

-0.64

JV

-0.64

POSITIVE LOGITS

 herself

1.76

 husband

1.03

 breasts

0.94

 boyfriend

0.93

 pregnant

0.93

 childbirth

0.88

 maternity

0.88

 maid

0.88

 menstru

0.87

 nurse

0.86

Activations Density 1.145%

personal interactions and conversations involving specific actions and emotions

instances of reported speech or claims made by individuals

No Comments

No Known Activations