INDEX

Explanations

mentions of specific assignments or tasks, particularly those involving some type of operation or objective

oai_token-act-pair · gpt-3.5-turbo

references to various "missions" within the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Column

-0.81

ãģį

-0.79

picking

-0.75

orian

-0.70

urrency

-0.68

urrencies

-0.68

OFF

-0.67

 Ange

-0.66

batch

-0.65

cling

-0.64

POSITIVE LOGITS

 mission

0.93

 missions

0.88

arily

0.83

 creep

0.82

 abroad

0.78

 Mission

0.76

eers

0.76

 accomplished

0.74

Mission

0.74

 Impossible

0.74

Activations Density 0.013%

mentions of specific assignments or tasks, particularly those involving some type of operation or objective

references to various "missions" within the text

No Comments

No Known Activations