INDEX

Explanations

terms related to organizations and projects, with a focus on specific roles and teams within them

oai_token-act-pair · gpt-3.5-turbo

references to organized groups or teams

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

Downloadha

-0.72

akespeare

-0.71

NetMessage

-0.67

iphate

-0.64

azo

-0.61

 Institutes

-0.60

cyclopedia

-0.60

ournaments

-0.60

 wildfires

-0.60

ï¸ı

-0.59

POSITIVE LOGITS

 manager

1.57

 leader

1.50

 chairman

1.50

 coordinator

1.49

 director

1.48

leader

1.41

 supervisor

1.40

mate

1.40

 chief

1.40

 president

1.40

Activations Density 0.307%

terms related to organizations and projects, with a focus on specific roles and teams within them

references to organized groups or teams

No Comments

No Known Activations