INDEX

Explanations

job titles related to management, particularly in a corporate or organizational context

oai_token-act-pair · gpt-3.5-turbo

references to various managerial roles or positions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

 canon

-0.67

 passage

-0.65

 parallel

-0.65

iny

-0.64

dot

-0.63

 testament

-0.62

 seas

-0.60

 republic

-0.60

proof

-0.60

 trem

-0.60

POSITIVE LOGITS

 Manager

3.91

 manager

2.32

manager

2.24

Manager

2.19

 managers

1.72

 Supervisor

1.68

 Coordinator

1.64

 Management

1.55

Management

1.51

 Director

1.48

Activations Density 0.014%

job titles related to management, particularly in a corporate or organizational context

references to various managerial roles or positions

No Comments

No Known Activations