INDEX

Explanations

programming syntax and structures

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

programming code and types

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

bindingNavigator

-0.46

 exercício

-0.44

Dichloropropane

-0.43

 Wirkung

-0.42

 dostęp

-0.41

 Zustimmung

-0.40

舺

-0.40

oblastoma

-0.40

 opdracht

-0.40

 murni

-0.40

POSITIVE LOGITS

MigrationBuilder

0.52

 Roskov

0.48

migrationBuilder

0.46

Попис

0.44

 Chwiliwch

0.43

 XNUMX

0.41

 MonoBehaviour

0.40

 ControllerBase

0.40

 الحره

0.40

তথ্যসূত্র

0.40

Activations Density 0.003%

programming syntax and structures

programming code and types

No Comments

No Known Activations