INDEX

Explanations

phrases indicating a comparison or contrast

oai_token-act-pair · gpt-3.5-turbo

language indicating contrast or opposition in discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

redo

-0.71

eda

-0.58

enegger

-0.57

 Highlights

-0.55

 Dresden

-0.54

mberg

-0.53

 Conquer

-0.53

 Semin

-0.53

Wag

-0.52

Bam

-0.52

POSITIVE LOGITS

to

0.92

 thereto

0.91

itably

0.71

thodox

0.70

acles

0.68

uitive

0.67

to

0.67

TO

0.64

ract

0.62

osite

0.60

Activations Density 0.019%

phrases indicating a comparison or contrast

language indicating contrast or opposition in discussions

No Comments

No Known Activations