INDEX

Explanations

commands written in a specific format

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Demokrat

-0.66

 OMIT

-0.61

WithMany

-0.59

 demokrat

-0.59

 kär

-0.57

Biografi

-0.57

CheckedChanged

-0.56

Campe

-0.55

 bakteri

-0.55

Pró

-0.54

POSITIVE LOGITS

 disreg

1.31

 scrat

1.30

 suscep

1.27

 affor

1.25

 nutella

1.25

 impra

1.21

 swarovski

1.21

 tupperware

1.21

 increa

1.21

 cushi

1.20

Activations Density 0.130%

commands written in a specific format

No Comments

No Known Activations