INDEX

Explanations

words related to negative or unpleasant situations

oai_token-act-pair · gpt-3.5-turbo

the prefix "uns," indicating negation or absence

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

SHIP

-0.75

 Grayson

-0.70

stanbul

-0.68

zzo

-0.68

 Mercury

-0.67

OPLE

-0.67

*/(

-0.66

tsky

-0.66

 Guardians

-0.66

anwhile

-0.63

POSITIVE LOGITS

uns

0.92

avour

0.90

rep

0.89

apon

0.81

oci

0.80

uitive

0.80

conv

0.79

heat

0.79

 concess

0.78

alted

0.76

Activations Density 0.006%

words related to negative or unpleasant situations

the prefix "uns," indicating negation or absence

No Comments

No Known Activations