INDEX

Explanations

actions related to communication, such as reporting, debating, and giving interviews

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

emen

-0.83

.)

-0.80

é¾į

-0.80

 nutshell

-0.77

Eastern

-0.74

 totality

-0.73

edge

-0.68

Tsu

-0.68

 fullest

-0.68

 Typh

-0.67

POSITIVE LOGITS

ModLoader

0.96

etc

0.89

agher

0.82

isSpecialOrderable

0.79

acebook

0.79

Downloadha

0.78

 captcha

0.77

":""},{"

0.76

ifully

0.70

hement

0.68

Activations Density 14.495%

actions related to communication, such as reporting, debating, and giving interviews

No Comments

No Known Activations