INDEX

Explanations

phrases related to software development, technical documentation, and commands

oai_token-act-pair · gpt-3.5-turbo

various technical terms and concepts related to headers and data processing

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

ilipp

-0.83

itan

-0.68

neau

-0.67

oya

-0.65

arrett

-0.63

bably

-0.63

ahoo

-0.62

istine

-0.61

uy

-0.61

ttle

-0.61

POSITIVE LOGITS

 Provides

0.80

Eas

0.79

 Allows

0.76

 Expend

0.73

 Located

0.69

TBD

0.67

 Cosponsors

0.66

 Increases

0.66

 Contains

0.66

Allows

0.64

Activations Density 0.216%

phrases related to software development, technical documentation, and commands

various technical terms and concepts related to headers and data processing

No Comments

No Known Activations