INDEX

Explanations

terms related to skill levels and upgrades in a training or professional context

oai_token-act-pair · gpt-3.5-turbo

various levels of proficiency or categorization in skills or products

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

conservancy

-0.90

rael

-0.78

incent

-0.77

 ostensibly

-0.76

 supposedly

-0.72

aying

-0.69

 purportedly

-0.69

omething

-0.68

 contag

-0.68

orship

-0.68

POSITIVE LOGITS

 Mode

1.44

 Version

1.37

 Edition

1.37

 Selection

1.36

 Modes

1.29

 Definition

1.27

 Format

1.27

 Characters

1.26

 Release

1.24

 Distance

1.22

Activations Density 0.265%

terms related to skill levels and upgrades in a training or professional context

various levels of proficiency or categorization in skills or products

No Comments

No Known Activations