INDEX

Explanations

phrases indicating comparison or distinction between two entities

oai_token-act-pair · gpt-3.5-turbo

pairings or comparisons involving the number two

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

phies

-0.67

uel

-0.61

 Shutterstock

-0.60

mun

-0.60

uck

-0.59

Own

-0.58

 Thick

-0.57

ucks

-0.56

park

-0.55

employment

-0.55

POSITIVE LOGITS

 halves

1.05

 apiece

0.83

 respectively

0.69

 identical

0.69

 evenly

0.67

 sexes

0.63

ridges

0.62

ifles

0.62

 equally

0.60

 thirds

0.60

Activations Density 0.156%

phrases indicating comparison or distinction between two entities

pairings or comparisons involving the number two

No Comments

No Known Activations