INDEX

Explanations

words related to legal and investigative activities

oai_token-act-pair · gpt-3.5-turbo

references to arrests and legal investigations

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

 Moose

-0.71

 Parsons

-0.71

Weiss

-0.65

 Americas

-0.63

 Walt

-0.63

 Latino

-0.63

 Sounders

-0.62

 Diablo

-0.61

 Tacoma

-0.61

amer

-0.61

POSITIVE LOGITS

 till

1.36

 Gujarat

1.26

 Delhi

1.21

 Maharashtra

1.21

 Kerala

1.20

Rs

1.19

jriwal

1.16

hra

1.16

 Uttar

1.15

 lakh

1.15

Activations Density 0.687%

words related to legal and investigative activities

references to arrests and legal investigations

No Comments

No Known Activations