INDEX

Explanations

instances of the word 'product'

oai_token-act-pair · gpt-3.5-turbo

references to products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

aft

-0.67

 Shore

-0.63

erved

-0.62

orld

-0.62

 Cavern

-0.61

FIN

-0.61

Lod

-0.61

 Tavern

-0.61

 Deck

-0.60

den

-0.60

POSITIVE LOGITS

 product

1.01

Products

0.99

 products

0.97

Product

0.93

ivity

0.92

Hunt

0.84

ively

0.80

ivities

0.78

 manufact

0.78

Interstitial

0.78

Activations Density 0.020%

instances of the word 'product'

references to products

No Comments

No Known Activations