INDEX

Explanations

products, stores, and shopping-related terms

oai_token-act-pair · gpt-3.5-turbo

references to stores or shopping-related activities

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

Ö¼

-0.73

èĢħ

-0.70

udic

-0.70

 prol

-0.69

emonic

-0.66

illian

-0.65

itely

-0.61

azeera

-0.61

 imperative

-0.61

obin

-0.61

POSITIVE LOGITS

 shelves

0.69

bats

0.67

 Auction

0.67

 Stores

0.64

taboola

0.62

 Discount

0.62

wagen

0.61

 hamb

0.60

seys

0.60

 Sears

0.59

Activations Density 0.984%

products, stores, and shopping-related terms

references to stores or shopping-related activities

No Comments

No Known Activations