INDEX

Explanations

mentions of law enforcement entities and actions

oai_token-act-pair · gpt-3.5-turbo

terms related to enforcement and regulatory actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

gets

-0.80

ranged

-0.79

php

-0.78

Mem

-0.76

bian

-0.76

Boss

-0.74

mem

-0.73

ones

-0.70

range

-0.69

Ã§

-0.69

POSITIVE LOGITS

 enforcement

1.18

 Enforcement

1.17

 crackdown

0.95

 sanction

0.91

 undermin

0.90

 poaching

0.85

 conflic

0.83

enforcement

0.83

 estab

0.82

 tampering

0.80

Activations Density 0.010%

mentions of law enforcement entities and actions

terms related to enforcement and regulatory actions

No Comments

No Known Activations