INDEX

Explanations

phrases related to taking action or making plans

oai_token-act-pair · gpt-3.5-turbo

actions related to structuring, organizing, or categorizing items or concepts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

terness

-0.68

udence

-0.67

imei

-0.65

edom

-0.65

plom

-0.65

perty

-0.64

xxx

-0.63

uden

-0.62

idia

-0.61

ylon

-0.61

POSITIVE LOGITS

matically

0.87

uate

0.84

 oneself

0.83

 ourselves

0.80

ulate

0.73

 Flare

0.71

 Statements

0.71

ħ

0.69

 yourself

0.67

Ī

0.67

Activations Density 0.261%

phrases related to taking action or making plans

actions related to structuring, organizing, or categorizing items or concepts

No Comments

No Known Activations