INDEX

Explanations

technical terms and concepts related to analysis, particularly in a statistical or computational context

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

mathematical expressions and abstract 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 surla

-0.35

 informée

-0.33

AutoScaleMode

-0.33

 муніципалі

-0.30

 critical

-0.28

homonymie

-0.28

 milioane

-0.27

控

-0.27

ละ

-0.27

以

-0.27

POSITIVE LOGITS

الدراسه

0.66

 Pardavimas

0.58

 StatelessWidget

0.57

ésultats

0.56

⸄

0.56

<unused43>

0.56

<unused52>

0.56

<unused14>

0.56

[@BOS@]

0.56

<unused3>

0.56

Activations Density 0.000%

technical terms and concepts related to analysis, particularly in a statistical or computational context

mathematical expressions and abstract concepts

No Comments

No Known Activations