INDEX

Explanations

references to duality or things that exist in pairs

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

both ends or sides

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

AlterField

-0.82

ագրություններ

-0.80

findpost

-0.79

 كومونز

-0.76

MessageTagHelper

-0.75

 nonUne

-0.74

TryDecodeAsNil

-0.74

VersionUID

-0.73

Personensuche

-0.73

Hozzáférés

-0.73

POSITIVE LOGITS

 résister

0.32

 adopción

0.26

 visuel

0.24

 administrativo

0.24

 gleicher

0.24

NotIn

0.23

 Europese

0.23

 historical

0.22

 ocasi

0.21

 liné

0.21

Activations Density 0.027%

references to duality or things that exist in pairs

both ends or sides

No Comments

No Known Activations