INDEX

Explanations

terms related to law enforcement

oai_token-act-pair · gpt-3.5-turbo

references to law enforcement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

xit

-0.75

 Carbuncle

-0.72

 Philips

-0.67

 Cinnamon

-0.66

 Tate

-0.64

Dak

-0.64

Voy

-0.62

 Shib

-0.61

 Medium

-0.60

 Gree

-0.60

POSITIVE LOGITS

 enforcement

1.40

suit

1.19

 Enforcement

1.13

enforcement

1.12

makers

1.10

 abiding

1.07

suits

1.07

lessness

1.04

breakers

1.02

yer

1.01

Activations Density 0.042%

terms related to law enforcement

references to law enforcement

No Comments

No Known Activations