INDEX

Explanations

expressions related to numerical values

oai_token-act-pair · gpt-3.5-turbo

phrases that convey positive numerical quantities or statistical information

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

The tokens " plus" "plus" "minus" " minus"

Explanation Uploaded by User

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

abies

-0.75

dfx

-0.74

ugu

-0.74

 Sparkle

-0.70

ruary

-0.70

stem

-0.70

adr

-0.69

wolf

-0.68

bris

-0.67

ollar

-0.67

POSITIVE LOGITS

cules

1.09

minus

0.78

/-

0.77

ï¸

0.73

++++

0.72

-+-+-+-+

0.67

 Scotia

0.67

henko

0.63

lihood

0.63

 infinity

0.63

Activations Density 0.034%

expressions related to numerical values

phrases that convey positive numerical quantities or statistical information

The tokens " plus" "plus" "minus" " minus"

No Comments

No Known Activations