INDEX

Explanations

terms related to military conflicts and politics, particularly focusing on the wars in Iraq and Afghanistan

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

rawdownloadcloneembedreportprint

-0.77

atin

-0.73

onym

-0.70

ateur

-0.66

 antioxid

-0.65

pressed

-0.63

 newcom

-0.62

bery

-0.62

Lear

-0.61

reci

-0.60

POSITIVE LOGITS

 ravaged

0.76

 Syndrome

0.72

 Basin

0.72

 annex

0.71

ghazi

0.69

 Hussein

0.69

 escalated

0.66

 devastated

0.66

 spir

0.65

 Reconstruction

0.64

Activations Density 10.407%

terms related to military conflicts and politics, particularly focusing on the wars in Iraq and Afghanistan

No Comments

No Known Activations