INDEX

Explanations

phrases related to challenging, questioning, or undermining various systems, laws, or institutions

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

 renfer

-0.46

Prova

-0.46

Consig

-0.45

Ņ

-0.44

UIButton

-0.43

Fails

-0.43

 fraudul

-0.42

 désigne

-0.42

marzo

-0.42

 garantit

-0.42

POSITIVE LOGITS

 osal

1.00

 territo

0.85

 jaya

0.83

 naer

0.83

 zove

0.82

haviour

0.81

 saad

0.79

 Minang

0.79

 chèvre

0.78

 maksi

0.78

Activations Density 0.390%

phrases related to challenging, questioning, or undermining various systems, laws, or institutions

No Comments

No Known Activations