INDEX

Explanations

words related to political conflict and focus on specific topics or issues

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

 nuisance

-0.68

 bigot

-0.66

 dentist

-0.65

 Pharm

-0.65

 Delicious

-0.64

 trump

-0.63

 Conquer

-0.62

 Clown

-0.62

 begg

-0.62

 photoc

-0.62

POSITIVE LOGITS

 blame

1.11

vre

1.07

 emphasis

0.96

phasis

0.90

 attention

0.89

 focus

0.88

ategory

0.86

staff

0.85

 Attention

0.84

rescent

0.83

Activations Density 16.472%

words related to political conflict and focus on specific topics or issues

No Comments

No Known Activations