INDEX

Explanations

words related to "Active Directory" and technological terms

oai_token-act-pair · gpt-3.5-turbo

references to "Active" states or statuses, particularly within technological or organizational contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

OUGH

-0.83

ĸļ

-0.81

bill

-0.75

Ń·

-0.71

awed

-0.68

pper

-0.67

udging

-0.66

aneers

-0.64

 Haram

-0.64

 Barnett

-0.64

POSITIVE LOGITS

 Active

1.03

 Directory

0.98

Active

0.95

ives

0.86

ived

0.83

active

0.80

tymology

0.77

igue

0.77

ive

0.76

 Passive

0.72

Activations Density 0.004%

words related to "Active Directory" and technological terms

references to "Active" states or statuses, particularly within technological or organizational contexts

No Comments

No Known Activations