INDEX

Explanations

information related to current events or news stories

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

isively

-0.70

ngth

-0.65

cius

-0.64

izen

-0.61

pick

-0.60

zai

-0.56

negie

-0.56

 slate

-0.56

farious

-0.56

 Travels

-0.55

POSITIVE LOGITS

 disapp

0.65

 happening

0.62

ECA

0.59

Ãĥ

0.59

 Shea

0.58

aiden

0.56

oggle

0.56

 Happ

0.56

rue

0.55

 happ

0.55

Activations Density 5.604%

information related to current events or news stories

No Comments

No Known Activations