INDEX

Explanations

mathematical expressions containing addition

oai_token-act-pair · gpt-3.5-turbo

expressions related to scoring or adding values

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Runner

-0.69

 Grimes

-0.68

soDeliveryDate

-0.67

 Nare

-0.66

 Dise

-0.66

 Sparkle

-0.64

pace

-0.63

opathic

-0.63

agra

-0.63

 Petty

-0.63

POSITIVE LOGITS

-+-+-+-+

1.28

++++++++++++++++

1.24

/+

1.07

++++

0.97

-+

0.92

--+

0.91

++++++++

0.91

/-

0.86

-+-+

0.76

0.71

Activations Density 0.023%

mathematical expressions containing addition

expressions related to scoring or adding values

No Comments

No Known Activations