INDEX

Explanations

mentions of specific entertainment franchises and media titles

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

hers

-0.67

bat

-0.64

rn

-0.63

occ

-0.62

omorph

-0.61

ible

-0.61

het

-0.61

ele

-0.60

 enrol

-0.60

hel

-0.59

POSITIVE LOGITS

etheless

0.96

 Citiz

0.89

akeru

0.71

 Paste

0.69

senal

0.66

 Nanto

0.64

 Nept

0.63

XVI

0.63

 Fernandez

0.63

 spons

0.63

Activations Density 9.110%

mentions of specific entertainment franchises and media titles

No Comments

No Known Activations