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. 22-GEMMASCOPE-RES-16K
    6. 10321
    Prev
    Next
    INDEX
    Explanations

    references to Azerbaijani culture and musical instruments

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

    Follows punctuation or symbols

    np_acts-logits-general · gemini-2.0-flash

    preceding specific tokens

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.hook_resid_post
    Hook Layer
    22
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ghijkl
    -0.44
    asInstanceOf
    -0.44
    addGap
    -0.44
    stateProvider
    -0.44
     daly
    -0.43
     mait
    -0.42
    oung
    -0.42
    Pemer
    -0.42
    ather
    -0.42
    Xna
    -0.41
    POSITIVE LOGITS
    ThroughAttribute
    0.81
    TypedDataSet
    0.68
    contentLoaded
    0.65
     və
    0.63
     Azerbaijan
    0.61
    ə
    0.60
    maq
    0.60
     Roskov
    0.60
     createSlice
    0.60
    awtextra
    0.60
    Activations Density 0.012%

    No Known Activations