INDEX

Explanations

concepts related to categorization and assessment

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

concepts and states

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

AndEndTag

-0.65

principalColumn

-0.57

+#+#

-0.55

Smokey

-0.53

 Smokey

-0.53

ocardio

-0.51

AddTagHelper

-0.51

 disambiguazione

-0.49

 つ折り

-0.49

mobileqq

-0.49

POSITIVE LOGITS

المكان

0.38

IndentedString

0.33

meaning

0.29

hmigung

0.29

 meaning

0.28

️

0.28

 définition

0.27

 hardware

0.27

 Что

0.27

 Eigentümer

0.27

Activations Density 0.236%

concepts related to categorization and assessment

concepts and states

No Comments

No Known Activations