INDEX

Explanations

phrases related to law enforcement agencies, operations, and activities

oai_token-act-pair · gpt-3.5-turbo

references to law enforcement agencies and their activities

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

 splits

-0.74

 bush

-0.71

 sink

-0.70

 burn

-0.69

 impressions

-0.69

 pasta

-0.68

 cheaply

-0.66

 split

-0.66

cup

-0.65

 stab

-0.65

POSITIVE LOGITS

 Enforcement

3.94

enforcement

1.69

 Investigations

1.34

 enforcement

1.28

 Compliance

1.25

 Investigative

1.15

 Regulatory

1.14

 Prevention

1.11

 Examination

1.08

 Investig

1.08

Activations Density 0.011%

phrases related to law enforcement agencies, operations, and activities

references to law enforcement agencies and their activities

No Comments

No Known Activations