INDEX

Explanations

information related to news articles and political events

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

SPONSORED

-0.83

Newsletter

-0.81

Secondly

-0.77

}.

-0.77

âĶĢâĶĢâĶĢâĶĢ

-0.71

phasis

-0.71

åĤ

-0.71

ļéĨĴ

-0.70

 Secondly

-0.70

âĹ¼

-0.69

POSITIVE LOGITS

 watchdog

0.72

 gunman

0.69

zbollah

0.64

 POLITICO

0.64

 awoke

0.63

 embattled

0.62

reetings

0.62

"))

0.61

heed

0.61

 bicycl

0.61

Activations Density 22.878%

information related to news articles and political events

No Comments

No Known Activations