INDEX

Explanations

references to mathematical or analytical relationships and comparisons in scientific contexts

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

characteristic vs functions

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

WithMany

-0.50

Становништво

-0.42

 становника

-0.42

 IBOutlet

-0.42

nimo

-0.41

quehanna

-0.41

mandu

-0.40

fxml

-0.40

especially

-0.40

kkue

-0.40

POSITIVE LOGITS

GEBURTSDATUM

0.55

KURZBESCHREIBUNG

0.46

 Autorisations

0.44

MigrationBuilder

0.42

extAlignment

0.38

באנגלית

0.35

CppCodeGen

0.35

 defaultstate

0.35

 Jungs

0.35

ChildScrollView

0.35

Activations Density 0.738%

references to mathematical or analytical relationships and comparisons in scientific contexts

characteristic vs functions

No Comments

No Known Activations