INDEX

Explanations

answers to various questions or statements

oai_token-act-pair · gpt-3.5-turbo

references to answers and responses to questions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

chin

-0.76

wana

-0.69

ammy

-0.69

DAQ

-0.66

eaturing

-0.64

ju

-0.63

akin

-0.63

outhern

-0.62

eatures

-0.62

uj

-0.62

POSITIVE LOGITS

 answ

1.03

 thereto

0.94

answered

0.92

 answered

0.91

yes

0.91

 answers

0.90

 questions

0.90

 answer

0.89

naires

0.86

answer

0.85

Activations Density 0.040%

answers to various questions or statements

references to answers and responses to questions

No Comments

No Known Activations