INDEX

Explanations

values and measurements presented in specific formats with numbers and units

oai_token-act-pair · gpt-3.5-turbo

numerical ratings or scores related to content

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

Wallet

-0.73

 codes

-0.66

PDF

-0.64

soDeliveryDate

-0.62

 culminated

-0.61

trace

-0.59

Torrent

-0.59

prototype

-0.58

 coordinates

-0.58

pad

-0.57

POSITIVE LOGITS

[+

0.85

ã

0.77

 compliment

0.70

 Respect

0.68

 compliments

0.67

 Debor

0.64

 sarc

0.63

cake

0.62

efe

0.61

 Cosponsors

0.61

Activations Density 0.041%

values and measurements presented in specific formats with numbers and units

numerical ratings or scores related to content

No Comments

No Known Activations