INDEX

Explanations

the word "smirk" and related variations

oai_token-act-pair · gpt-3.5-turbo

instances of the prefix "sm" 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 @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

DragonMagazine

-0.83

hyde

-0.68

 Vanguard

-0.65

Syl

-0.62

 Emir

-0.62

EMENT

-0.62

quickShipAvailable

-0.61

UFC

-0.61

 Ecuador

-0.60

 è£ıè

-0.60

POSITIVE LOGITS

itten

1.20

attering

1.18

ithing

1.13

oky

1.13

ugg

1.09

ashing

1.07

ashed

1.04

udge

1.04

ould

1.03

acking

1.03

Activations Density 0.015%

the word "smirk" and related variations

instances of the prefix "sm" in various contexts

No Comments

No Known Activations