INDEX

Explanations

specific mentions of news outlets or organizations

oai_token-act-pair · gpt-3.5-turbo

references to organizations or entities related to investigative or journalistic reporting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

lessly

-0.77

 convincing

-0.76

 tune

-0.71

 domestically

-0.71

 chains

-0.70

 behold

-0.70

 reel

-0.68

 trump

-0.68

 orchestra

-0.68

 cheaply

-0.68

POSITIVE LOGITS

 Affairs

1.40

 Warfare

1.27

 Investigations

1.25

 Advisor

1.25

 Assistance

1.25

 Relations

1.24

 Assessment

1.23

 Development

1.22

 Adviser

1.22

 Services

1.21

Activations Density 0.162%

specific mentions of news outlets or organizations

references to organizations or entities related to investigative or journalistic reporting

No Comments

No Known Activations