INDEX

Explanations

proper nouns and noun phrases related to politics, government, and societal 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

shell

-0.51

Args

-0.49

iP

-0.49

versions

-0.48

 accompan

-0.48

 shells

-0.47

pus

-0.47

charges

-0.47

activity

-0.46

 arrivals

-0.45

POSITIVE LOGITS

 Lauder

0.66

child

0.60

ocracy

0.58

hips

0.58

 gracious

0.58

 Reserved

0.57

osate

0.56

 Mutual

0.56

eous

0.55

rats

0.55

Activations Density 15.912%

proper nouns and noun phrases related to politics, government, and societal issues

No Comments

No Known Activations