INDEX

Explanations

instances where text mentions an excessive amount or intensity of something

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

byn

-0.86

saf

-0.83

selves

-0.82

Ń·

-0.81

stead

-0.79

İĭ

-0.77

ands

-0.77

iversary

-0.73

psons

-0.73

sylv

-0.73

POSITIVE LOGITS

 emphasis

0.92

 inconsistency

0.86

 attention

0.85

inventoryQuantity

0.82

 trouble

0.82

 effort

0.81

 hassle

0.80

 hype

0.79

 temptation

0.78

dmg

0.76

Activations Density 10.640%

instances where text mentions an excessive amount or intensity of something

No Comments

No Known Activations