INDEX

Explanations

phrases indicating safety or recommendations for improvement

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

expandindo

-0.70

istert

-0.67

addCriterion

-0.65

RUnlock

-0.59

ingeki

-0.57

Mayor

-0.57

epam

-0.56

::_('

-0.54

levering

-0.54

hematical

-0.54

POSITIVE LOGITS

 estekak

0.68

 lieber

0.63

use

0.62

 préférable

0.60

你不要

0.57

干脆

0.57

 préfère

0.56

 Besten

0.55

 tartalomajánló

0.54

你先

0.53

Activations Density 0.215%

phrases indicating safety or recommendations for improvement

No Comments

No Known Activations

phrases indicating safety or recommendations for improvement

No Comments

No Known Activations