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. 5631
    Prev
    Next
    INDEX
    Explanations

    action verbs related to competition and sports events

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

    Followed by "on", "out", or "back"

    np_acts-logits-general · gemini-2.0-flash

    verbs of motion followed by direction

    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
     ThemeData
    -0.51
    ScreenState
    -0.49
    ModelState
    -0.49
    ByUsername
    -0.49
    weile
    -0.48
    writeField
    -0.48
    例句
    -0.48
    makeConstraints
    -0.47
     للاسماء
    -0.46
    regen
    -0.46
    POSITIVE LOGITS
    
    0.70
     disambiguazione
    0.69
    homonymie
    0.63
    //
    0.60
    ⓧ
    0.57
    COUVER
    0.56
    脚注の使い方
    0.55
    HasColumnType
    0.54
    ThroughAttribute
    0.54
    errHandler
    0.52
    Activations Density 0.103%

    No Known Activations