INDEX

Explanations

statements that accuse or assert blame on someone or something

oai_token-act-pair · gpt-3.5-turbo

instances of the word "def" and its various forms related to defamation or accusations

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

sth

-0.71

 attachments

-0.69

 Madness

-0.67

 terminals

-0.62

QB

-0.61

hyde

-0.59

 Boll

-0.58

natureconservancy

-0.58

 terminal

-0.57

DAY

-0.57

POSITIVE LOGITS

erence

1.28

ensible

1.28

ection

1.27

ector

1.23

lated

1.22

lection

1.22

acement

1.20

ected

1.19

lected

1.17

amiliar

1.16

Activations Density 0.013%

statements that accuse or assert blame on someone or something

instances of the word "def" and its various forms related to defamation or accusations

No Comments

No Known Activations