INDEX

Explanations

phrases related to technical processes and structures

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

 Healthcare

-0.53

ilitary

-0.53

 PsyNet

-0.52

ively

-0.49

licts

-0.49

 Commit

-0.48

 Missions

-0.48

aeda

-0.48

arios

-0.47

 Finance

-0.46

POSITIVE LOGITS

 itself

0.57

 pictured

0.56

washer

0.55

face

0.55

 analogy

0.53

 behaves

0.53

wright

0.53

powder

0.53

 diameter

0.53

pole

0.52

Activations Density 21.415%

phrases related to technical processes and structures

No Comments

No Known Activations