© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW
    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 3066
    Prev
    Next
    INDEX
    Explanations

    phrases indicating judgments or assessments about actions and individuals

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    end punctuation followed by sentence starters

    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
    20
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     kasarigan
    -0.45
    nsito
    -0.44
    ISupport
    -0.44
    专属
    -0.40
     Issus
    -0.36
     başarı
    -0.36
     laboratoire
    -0.36
     labs
    -0.35
     Unidas
    -0.35
     avoient
    -0.35
    POSITIVE LOGITS
     autorytatywna
    0.52
    ViewFeatures
    0.48
    PhysRevD
    0.46
     saites
    0.45
     disappointed
    0.45
     eulogy
    0.42
    DBNull
    0.42
    断
    0.41
     regret
    0.40
     HasFactory
    0.40
    Activations Density 0.065%

    No Known Activations