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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 120203
    Prev
    Next
    INDEX
    Explanations

    terms related to mathematical functions and their representations

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

    sigmoid infectivity function

    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
    __':
    
    -0.83
    EndInit
    -0.75
    tvguidetime
    -0.74
    <unused41>
    -0.74
    <unused23>
    -0.73
    <unused17>
    -0.73
    <unused28>
    -0.73
    <unused74>
    -0.73
    <unused79>
    -0.73
    [@BOS@]
    -0.73
    POSITIVE LOGITS
     riesgos
    0.35
     tebal
    0.35
    sigmoid
    0.35
     Fonction
    0.34
     Fungsi
    0.34
    function
    0.33
     funcion
    0.33
     hermos
    0.32
     function
    0.32
     quema
    0.32
    Activations Density 0.040%

    No Known Activations