INDEX

Explanations

titles, names, or phrases starting with "Do"

oai_token-act-pair · gpt-3.5-turbo

instances of the word "do" and its variations in different contexts

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

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

Reviewer

-0.76

)=(

-0.74

workshop

-0.71

boarding

-0.71

ONSORED

-0.69

ulates

-0.62

ItemTracker

-0.62

mus

-0.61

tnc

-0.61

ulative

-0.61

POSITIVE LOGITS

omsday

1.31

herty

1.25

ppel

1.15

zens

1.13

lez

1.08

ctr

1.03

gging

0.99

ctors

0.98

oley

0.95

berman

0.93

Activations Density 0.086%

titles, names, or phrases starting with "Do"

instances of the word "do" and its variations in different contexts

No Comments

No Known Activations