INDEX

Explanations

names containing the letters "Ra" followed by a numerical or alphabetical value

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

llah

-0.65

ERAL

-0.64

 Phelps

-0.64

 Merchants

-0.63

intosh

-0.63

 à¨

-0.62

é¾įå¥ĳå£«

-0.62

 EntityItem

-0.62

 Idle

-0.61

rary

-0.61

POSITIVE LOGITS

isin

1.33

fters

1.16

ven

1.14

ving

1.14

vel

1.13

fter

1.12

isen

1.12

ises

1.09

uth

1.07

ishi

1.04

Activations Density 5.987%

names containing the letters "Ra" followed by a numerical or alphabetical value

No Comments

No Known Activations