INDEX

Explanations

mentions of locations or people named "Mal" with varying levels of importance

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

SPONSORED

-0.58

 dunno

-0.58

guide

-0.57

 Revenge

-0.56

akura

-0.54

 Folder

-0.53

BIL

-0.52

 aroma

-0.52

 regard

-0.52

 fray

-0.52

POSITIVE LOGITS

nikov

0.97

mson

0.93

steen

0.84

coni

0.81

ochem

0.78

imov

0.75

Ń·

0.74

acus

0.73

ornia

0.72

imaru

0.72

Activations Density 0.213%

mentions of locations or people named "Mal" with varying levels of importance

proper nouns, particularly names

No Comments

No Known Activations