INDEX

Explanations

references to historical wars, especially World War II

oai_token-act-pair · gpt-3.5-turbo

mentions of World Wars, especially World War II

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

*/(

-0.86

obook

-0.84

ITY

-0.74

uckland

-0.73

66666666

-0.70

OGR

-0.69

ITIES

-0.69

aminer

-0.65

METHOD

-0.63

Asset

-0.62

POSITIVE LOGITS

lord

0.98

lords

0.94

riors

0.90

rior

0.89

fare

0.84

 Patton

0.81

fighter

0.80

 Syndrome

0.77

Osw

0.76

Vet

0.75

Activations Density 0.016%

references to historical wars, especially World War II

mentions of World Wars, especially World War II

No Comments

No Known Activations