INDEX

Explanations

references to onboard diagnostic systems and related technical terms

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

technical component analysis

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

 deseado

-0.35

τ

-0.34

-0.32

 Rode

-0.32

 packed

-0.31

 provided

-0.31

 corda

-0.31

 quedado

-0.30

 skill

-0.30

 distinguished

-0.30

POSITIVE LOGITS

GraphicsUnit

0.79

 Италијани

0.72

 Мексичка

0.68

󠁴

0.67

 ſche

0.67

 myſelf

0.65



0.64

MessageTagHelper

0.61

balleur

0.61

 disambiguazione

0.61

Activations Density 0.005%

references to onboard diagnostic systems and related technical terms

technical component analysis

No Comments

No Known Activations