INDEX

Explanations

phrases related to actions being taken or performed by different entities

oai_token-act-pair · gpt-3.5-turbo

the word "as" used in various 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

go

-0.63

stals

-0.62

tty

-0.60

)))

-0.60

âĹ¼

-0.57

æĸ

-0.57

 Peak

-0.57

roads

-0.56

eur

-0.56

FACE

-0.56

POSITIVE LOGITS

 evidenced

1.31

bestos

1.11

 opposed

1.04

semb

0.99

 shown

0.95

piration

0.90

piring

0.90

 well

0.88

ylum

0.86

pired

0.85

Activations Density 0.091%

phrases related to actions being taken or performed by different entities

the word "as" used in various contexts

No Comments

No Known Activations