INDEX

Explanations

references to or actions related to taking pictures

oai_token-act-pair · gpt-3.5-turbo

terms related to photography and taking pictures

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

catentry

-0.75

ĪĴ

-0.74

Beg

-0.69

inventoryQuantity

-0.67

ifted

-0.66

coal

-0.65

argon

-0.65

hler

-0.64

uitive

-0.63

margin

-0.63

POSITIVE LOGITS

 VIDEOS

0.80

icles

0.68

Fax

0.67

 WATCHED

0.66

icle

0.66

 photograp

0.66

":-

0.66

photos

0.65

 Beam

0.64

 JPEG

0.63

Activations Density 0.035%

references to or actions related to taking pictures

terms related to photography and taking pictures

No Comments

No Known Activations