INDEX

Explanations

instances where something is being stored in a physical location

oai_token-act-pair · gpt-3.5-turbo

phrases indicating the presence of stored items or concealed subjects

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

 tempted

-0.72

redo

-0.71

catentry

-0.70

entimes

-0.68

etheless

-0.67

caster

-0.66

ideo

-0.65

ŀ

-0.64

voice

-0.64

 alike

-0.63

POSITIVE LOGITS

 lieu

1.11

 accordance

1.03

 limbo

1.00

 museums

0.92

animate

0.91

 caves

0.91

 situ

0.90

 warehouses

0.90

 toilets

0.90

 hotels

0.90

Activations Density 0.196%

instances where something is being stored in a physical location

phrases indicating the presence of stored items or concealed subjects

No Comments

No Known Activations