INDEX

Explanations

numerical values mixed with words related to the concept of progression or sequence

oai_token-act-pair · gpt-3.5-turbo

references to sequences or rankings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

mercial

-0.77

anwhile

-0.64

enegger

-0.60

glers

-0.57

 Frie

-0.54

ateurs

-0.54

opol

-0.50

allery

-0.49

ortunately

-0.48

rities

-0.48

POSITIVE LOGITS

 veto

0.49

oried

0.48

 connector

0.47

 verse

0.47

idian

0.47

eq

0.46

 parentheses

0.45

 coin

0.45

natureconservancy

0.45

iscopal

0.45

Activations Density 1.841%

numerical values mixed with words related to the concept of progression or sequence

references to sequences or rankings

No Comments

No Known Activations