INDEX

Explanations

instances of emphasis or certainty

oai_token-act-pair · gpt-3.5-turbo

affirmative statements or expressions of certainty

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ufact

-0.73

mits

-0.68

umbn

-0.68

Reviewer

-0.66

roups

-0.65

lings

-0.64

OTO

-0.63

ritch

-0.63

Hol

-0.62

lore

-0.62

POSITIVE LOGITS

 benefited

0.73

ãĥ´ãĤ¡

0.71

 defin

0.71

 qualifies

0.70

 detract

0.69

olkien

0.68

heon

0.68

 notor

0.67

 benef

0.67

qus

0.67

Activations Density 0.008%

instances of emphasis or certainty

affirmative statements or expressions of certainty

No Comments

No Known Activations