INDEX

Explanations

terms related to sexual harassment and misconduct, as well as accusations and actions associated with this behavior

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scl-ajt/6-res_scl-ajt

Prompts (Dashboard)

12,288 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

46,080

Data Type

torch.float32

Hook Point

blocks.6.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

apollo-research/Skylion007-openwebtext-tokenizer-gpt2

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

REL

-0.60

IR

-0.57

AY

-0.57

WT

-0.56

Dispatch

-0.55

 tune

-0.55

 Winds

-0.54

LD

-0.53

IVERS

-0.53

aceae

-0.53

POSITIVE LOGITS

 allegations

0.99

 accuser

0.84

 misconduct

0.78

 accusations

0.77

 allegation

0.77

 dossier

0.72

 grop

0.69

 grooming

0.69

 blackmail

0.68

 rape

0.68

Activations Density 11.601%

terms related to sexual harassment and misconduct, as well as accusations and actions associated with this behavior

No Comments

No Known Activations