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-2-9B-IT
    4. Residual Stream - 131k
    5. 20-GEMMASCOPE-RES-131K
    6. 2716
    Prev
    Next
    INDEX
    Explanations

    constructor method signatures in programming code

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

    initializing objects with initWith

    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
    RegressionTest
    -0.64
    batore
    -0.51
     desmotivaciones
    -0.50
     OMITBAD
    -0.50
    /*:
    -0.49
     estekak
    -0.49
    ௃
    -0.49
    MigrationBuilder
    -0.48
     ſou
    -0.48
    ஦
    -0.48
    POSITIVE LOGITS
     initWith
    1.83
    initWith
    1.59
     colorWith
    0.87
     numberWith
    0.78
    With
    0.69
     With
    0.67
    encodeWith
    0.66
    with
    0.63
     with
    0.56
     Witherspoon
    0.52
    Activations Density 0.003%

    No Known Activations