INDEX

Explanations

phrases related to legal or political matters and policy descriptions

oai_token-act-pair · gpt-3.5-turbo

New Auto-Interp

Configuration

neuronpedia/gpt2-small__res_scefr-ajt/6-res_scefr-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

aminer

-0.73

 tremend

-0.72

plin

-0.70

icter

-0.69

neck

-0.67

ritic

-0.65

edin

-0.63

rir

-0.63

nom

-0.60

 nowhere

-0.60

POSITIVE LOGITS

ocating

1.23

iances

1.20

ocation

1.11

iance

1.08

ocate

1.03

ergic

1.03

 sorts

1.03

 kinds

1.03

ocated

1.02

owing

0.94

Activations Density 0.939%

phrases related to legal or political matters and policy descriptions

No Comments

No Known Activations