INDEX

Explanations

names or variations of the name "Ke" followed by a digit, likely referring to specific individuals or characters

oai_token-act-pair · gpt-3.5-turbo

instances of the name "Ke"

oai_token-act-pair · gpt-3.5-turbo Triggered by @jordan

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

etheless

-0.81

ashtra

-0.78

åħī

-0.72

-0.71

IONS

-0.70

inators

-0.68

 guiActiveUnfocused

-0.68

Ö¼

-0.67

ION

-0.67

é¾įå¥ĳå£«

-0.67

POSITIVE LOGITS

ptic

1.11

pler

1.08

eling

1.07

SPA

1.07

bye

1.06

ck

1.03

lder

1.02

lem

1.01

pper

1.00

cker

0.97

Activations Density 7.411%

names or variations of the name "Ke" followed by a digit, likely referring to specific individuals or characters

instances of the name "Ke"

No Comments

No Known Activations