INDEX

Explanations

terms related to access permissions, system interaction, and methods of processing or preparation

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

technical terms following prepositions

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

onViewCreated

-0.57

ⓧ

-0.55

PYX

-0.52

 BorderSide

-0.52

gridx

-0.52

migrationBuilder

-0.49

usermodel

-0.44

 GenerationType

-0.44

tagext

-0.43

pasto

-0.43

POSITIVE LOGITS

 AssemblyTitle

0.47

épaisseur

0.45

Bibliograf

0.40

 AssemblyCulture

0.39

 Baumann

0.38

 esternos

0.36

 Sorg

0.36

enfance

0.35

.*")]

0.34

Scénario

0.33

Activations Density 0.089%

terms related to access permissions, system interaction, and methods of processing or preparation

technical terms following prepositions

No Comments

No Known Activations