INDEX

Explanations

locations or settings described as being surrounded by something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "surrounded."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

enberg

-0.75

omething

-0.66

urai

-0.65

 hindsight

-0.63

hart

-0.63

uh

-0.61

inen

-0.60

hoff

-0.60

ipeg

-0.59

notations

-0.58

POSITIVE LOGITS

by

0.75

 tightly

0.75

by

0.72

 walls

0.71

 perimeter

0.71

BY

0.68

 surrounded

0.67

htaking

0.64

estinal

0.63

Track

0.62

Activations Density 0.059%

locations or settings described as being surrounded by something

instances of the word "surrounded."

No Comments

No Known Activations