INDEX

Explanations

proper nouns related to political figures, events, and historical contexts

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

ailable

-0.70

PLIED

-0.63

ous

-0.59

ulhu

-0.58

ously

-0.58

generic

-0.57

ises

-0.57

ruck

-0.56

raints

-0.56

insula

-0.55

POSITIVE LOGITS

 Lyndon

0.66

Johnson

0.62

 ILCS

0.62

 Johnson

0.61

Bib

0.60

 tract

0.59

vard

0.59

 tracts

0.57

 Shoals

0.57

eston

0.57

Activations Density 8.330%

proper nouns related to political figures, events, and historical contexts

No Comments

No Known Activations