INDEX

Explanations

terms related to collaborative networks and recommendations

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

recommendation models

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

 שוליים

-0.66

 AssemblyCompany

-0.60

GraphicsUnit

-0.60

Билгалдахарш

-0.58

⟬

-0.57

saraba

-0.57

دانشنامهٔ

-0.57

Datuak

-0.57

MLLoader

-0.56

脚注の使い方

-0.56

POSITIVE LOGITS

Distribución

0.35

 culturel

0.32

ilaian

0.32

 susu

0.31

 recombination

0.31

 reward

0.30

 psychologie

0.29

Рекомен

0.29

 kult

0.29

 recommend

0.29

Activations Density 0.016%

terms related to collaborative networks and recommendations

recommendation models

No Comments

No Known Activations