INDEX

Explanations

references to ice cream-related words and phrases

oai_token-act-pair · gpt-3.5-turbo

references to ice cream

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

 latent

-0.69

builders

-0.67

ne

-0.66

bol

-0.65

ivot

-0.64

iosyn

-0.64

 Neph

-0.64

 Galile

-0.62

iets

-0.62

eport

-0.62

POSITIVE LOGITS

eteria

0.82

 cups

0.79

 vending

0.77

 Delivery

0.76

 Vendor

0.75

Cola

0.73

 commercials

0.73

 cake

0.73

 drink

0.72

 soda

0.72

Activations Density 0.082%

references to ice cream-related words and phrases

references to ice cream

No Comments

No Known Activations