INDEX

Explanations

code-related keywords and function calls for managing resources and client interactions in a programming context

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

List, Delete, New, Create actions

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

 Administrativna

-0.45

 homolog

-0.31

 Comprometido

-0.30

цездатний

-0.29

KommentareTeilen

-0.28

 fors

-0.28

oredCriteria

-0.28

 jdbcTemplate

-0.28

InvalidProtocol

-0.28

 sinte

-0.28

POSITIVE LOGITS

 незавершена

0.65

fortawesome

0.55

Personendaten

0.51

httphttps

0.51

المشاركات

0.50

blico

0.49

 vantagem

0.48

weightedMode

0.47

OptionsMenu

0.47

InitStruct

0.46

Activations Density 0.046%

code-related keywords and function calls for managing resources and client interactions in a programming context

List, Delete, New, Create actions

No Comments

No Known Activations