INDEX

Explanations

mentions of law enforcement actions and investigations

oai_token-act-pair · gpt-3.5-turbo

incidents involving inspections or warnings related to health and safety

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ankind

-0.64

çķ

-0.58

luster

-0.57

isSpecialOrderable

-0.55

 resurgence

-0.54

 inheritance

-0.54

Interstitial

-0.54

æĺ

-0.53

 unpop

-0.53

 unden

-0.53

POSITIVE LOGITS

 overseeing

0.74

Īè

0.65

 confidential

0.64

 oversees

0.64

 superv

0.64

 advising

0.63

 hired

0.62

 oversee

0.61

 discretion

0.60

 advise

0.60

Activations Density 1.515%

mentions of law enforcement actions and investigations

incidents involving inspections or warnings related to health and safety

No Comments

No Known Activations