INDEX

Explanations

references to historical events, specifically related to World War II

oai_token-act-pair · gpt-3.5-turbo

references to 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

ibilities

-0.96

EStream

-0.90

veyard

-0.88

ibly

-0.86

»Ĵ

-0.83

ibility

-0.82

Downloadha

-0.77

merce

-0.77

staking

-0.76

WARE

-0.76

POSITIVE LOGITS

 Wide

0.94

 Order

0.92

 Federation

0.91

 Championships

0.90

 Bank

0.90

 Trade

0.89

 Airways

0.88

 Heritage

0.85

Cup

0.85

 Series

0.83

Activations Density 0.026%

references to historical events, specifically related to World War II

references to World War II

No Comments

No Known Activations