Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    3. Gemma-2-2B
    4. Residual Stream - 16k
    5. 18-GEMMASCOPE-RES-16K
    6. 8679
    Prev
    Next
    INDEX
    Explanations

    characters with special abilities or powers

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

    preceding "matter" or "ak"

    np_acts-logits-general · gemini-2.0-flash

    Mardakko, Naga, Rakuten, Koguma

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     مشين
    -0.77
     Pyram
    -0.69
    adaptiveStyles
    -0.61
    stination
    -0.60
    setOnAction
    -0.56
     BoxFit
    -0.55
    forRoot
    -0.55
    ''');
    -0.54
    NewGuid
    -0.54
     Nelly
    -0.53
    POSITIVE LOGITS
     Dark
    0.54
    who
    0.52
     stadig
    0.51
    Dark
    0.49
     who
    0.49
     учё
    0.49
    loth
    0.48
     sorella
    0.48
     skak
    0.48
    Portail
    0.48
    Activations Density 0.359%

    No Known Activations