INDEX

Explanations

phrases related to completion or achievement

oai_token-act-pair · gpt-3.5-turbo

references to potential responsibilities and conditions related to actions or agreements

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

retty

-0.61

Haunted

-0.51

everal

-0.50

çīĪ

-0.50

VIDIA

-0.50

bnb

-0.48

uesday

-0.48

ench

-0.48

alach

-0.46

illery

-0.46

POSITIVE LOGITS

to

1.44

to

1.18

TO

1.00

To

0.94

To

0.93

 thereto

0.86

 toget

0.79

ta

0.77

TO

0.64

¿

0.62

Activations Density 0.513%

phrases related to completion or achievement

references to potential responsibilities and conditions related to actions or agreements

No Comments

No Known Activations