INDEX

Explanations

instances where someone is accused or suspected of committing various criminal acts

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 krab

-0.65

 kram

-0.64

 Ukraina

-0.60

 utop

-0.58

 besta

-0.58

eko

-0.57

 plak

-0.57

 Déf

-0.56

 conclud

-0.55

 optik

-0.55

POSITIVE LOGITS

 charge

1.38

 charged

1.33

 charges

1.31

charge

1.31

 Charge

1.30

 Charges

1.24

 charging

1.24

Charge

1.23

charged

1.19

charges

1.18

Activations Density 0.087%

instances where someone is accused or suspected of committing various criminal acts

No Comments

No Known Activations