INDEX

Explanations

phrases related to mechanisms or processes in scientific 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

iaries

-0.65

hiro

-0.64

minster

-0.62

gets

-0.59

atana

-0.59

uden

-0.59

ield

-0.56

 Parables

-0.55

oning

-0.55

hab

-0.54

POSITIVE LOGITS

 mechanism

0.81

 mechanisms

0.77

 whereby

0.76

utics

0.75

 Mechan

0.74

witz

0.68

ptoms

0.66

 workings

0.65

ality

0.63

 symmetry

0.63

Activations Density 11.207%

phrases related to mechanisms or processes in scientific contexts

No Comments

No Known Activations