INDEX

Explanations

instances of the word "hole" in texts

oai_token-act-pair · gpt-3.5-turbo

references to physical holes or gaps in various contexts

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

Bal

-0.73

cia

-0.72

NX

-0.71

 Riding

-0.65

ority

-0.65

ingham

-0.65

 livest

-0.63

NetMessage

-0.63

lux

-0.63

rul

-0.62

POSITIVE LOGITS

 drilled

1.25

 hole

1.24

 holes

1.18

dug

1.11

hole

1.09

 Hole

1.03

 punched

1.01

holes

0.95

pots

0.91

 poked

0.88

Activations Density 0.055%

instances of the word "hole" in texts

references to physical holes or gaps in various contexts

No Comments

No Known Activations