INDEX

Explanations

references to weapons and political controversies

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

anchester

-0.73

 dexter

-0.62

ceivable

-0.58

ients

-0.57

ctory

-0.57

 ancest

-0.57

ignt

-0.57

eme

-0.56

encies

-0.55

emies

-0.55

POSITIVE LOGITS

 namely

1.00

 Cosmos

0.71

ARM

0.69

viz

0.66

ARS

0.66

TEXTURE

0.61

0000000000000000

0.59

OGR

0.59

albeit

0.58

VK

0.57

Activations Density 13.312%

references to weapons and political controversies

No Comments

No Known Activations