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. 997
    Prev
    Next
    INDEX
    Explanations

    titles and positions related to legal and administrative roles

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

    "Assistant" followed by a professional title

    np_acts-logits-general · gemini-2.0-flash

    assistant and job titles

    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
    kennt
    -0.55
     дописавши
    -0.54
    ynki
    -0.52
     either
    -0.51
    tituzione
    -0.50
    DataField
    -0.48
    ︎
    -0.48
     Teres
    -0.48
    gebir
    -0.47
    jęt
    -0.47
    POSITIVE LOGITS
    TargetException
    0.79
    ///</
    0.76
    WithIOException
    0.72
     Efq
    0.72
    ⓧ
    0.72
    Personendaten
    0.72
    __":
    
    0.71
    sizeCache
    0.70
    -------------</
    0.70
    ActionCreators
    0.67
    Activations Density 0.071%

    No Known Activations