INDEX

Explanations

references to specific products or items, particularly kits or sets

oai_token-act-pair · gpt-3.5-turbo

references to various types of kits

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

arching

-0.81

 Deity

-0.75

 Democr

-0.69

 Observatory

-0.67

south

-0.67

ews

-0.66

ä¼

-0.65

 temptation

-0.65

åŃĲ

-0.63

encer

-0.62

POSITIVE LOGITS

chens

1.32

kit

1.03

 kits

0.98

kit

0.97

kat

0.94

chy

0.88

 Kats

0.82

etsu

0.81

chen

0.80

Kit

0.79

Activations Density 0.013%

references to specific products or items, particularly kits or sets

references to various types of kits

No Comments

No Known Activations