INDEX

Explanations

technical terms and code structures indicating functionality or widespread occurrences

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

widespread despite

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

UpDown

-0.37

djangoproject

-0.35

ніципалі

-0.35

 Bergmann

-0.35

JsonInclude

-0.34

ativit

-0.34

 الاحتلال

-0.33

UDAD

-0.33

 Thacker

-0.33

 Fleck

-0.33

POSITIVE LOGITS

greenrobot

0.76

Dao

0.58

Dao

0.54

gdx

0.52

EndContext

0.50

 noDo

0.48

Autoritní

0.48

dao

0.45

 للاسماء

0.45

 الحره

0.44

Activations Density 0.000%

technical terms and code structures indicating functionality or widespread occurrences

widespread despite

No Comments

No Known Activations