INDEX

Explanations

terms related to programming structures and configurations

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

paths and code separators

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

Personensuche

-0.62

ConstraintMaker

-0.58

 oprot

-0.54

 iNdEx

-0.54

httphttps

-0.53

новништво

-0.53

MigrationBuilder

-0.51

queryInterface

-0.50

DebuggerNonUser

-0.48

GenerationType

-0.47

POSITIVE LOGITS

KURZBESCHREIBUNG

0.36

impanan

0.35

hawatir

0.35

postsleuth

0.34

querque

0.34

///</

0.33

databind

0.33

Gog

0.33

ionales

0.33

ünst

0.31

Activations Density 0.002%

terms related to programming structures and configurations

paths and code separators

No Comments

No Known Activations