INDEX

Explanations

instances of individuals stepping into or out of positions or roles

oai_token-act-pair · gpt-3.5-turbo

instances of the word "stepped" 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 @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

anguages

-0.80

 Corpus

-0.76

 Tycoon

-0.72

olesc

-0.68

orem

-0.64

oil

-0.64

usage

-0.64

oro

-0.64

ertation

-0.63

abol

-0.63

POSITIVE LOGITS

DragonMagazine

0.89

 steps

0.85

 Steps

0.85

OHN

0.79

BuyableInstoreAndOnline

0.78

iblings

0.77

Ds

0.77

daughter

0.76

 forward

0.74

assy

0.72

Activations Density 0.020%

instances of individuals stepping into or out of positions or roles

instances of the word "stepped" in various contexts

No Comments

No Known Activations