INDEX

Explanations

mentions of physical holes

oai_token-act-pair · gpt-3.5-turbo

references to holes 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 @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

cia

-0.82

ullivan

-0.72

 livest

-0.69

NetMessage

-0.69

against

-0.68

Bal

-0.65

ione

-0.65

anche

-0.64

ggles

-0.64

 Garner

-0.63

POSITIVE LOGITS

 hole

1.27

 holes

1.22

 drilled

1.16

hole

1.15

 Hole

1.10

dug

1.02

holes

0.97

 punched

0.89

pots

0.85

 poked

0.84

Activations Density 0.049%

mentions of physical holes

references to holes in various contexts

No Comments

No Known Activations