INDEX

Explanations

keywords related to well-known religions and their concepts

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

domains and concepts

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 ujednoznacz

-0.61

 propOrder

-0.59

+#+#

-0.56

 NSCoder

-0.55

 Italijani

-0.54

Jeografia

-0.53

extAlignment

-0.53

GOTREF

-0.50

Blon

-0.48

angelog

-0.48

POSITIVE LOGITS

 utafitiHapana

0.47

 yeux

0.35

 &___

0.31

 okuyayım

0.29

 بام

0.29

]-->

0.28

 corações

0.28

 متحده

0.27

 vectorielle

0.27

 oamen

0.27

Activations Density 0.047%

keywords related to well-known religions and their concepts

domains and concepts

No Comments

No Known Activations