© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    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. 42052
    Prev
    Next
    INDEX
    Explanations

    referenced whips and related actions or descriptions involving them

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

    whipping cream, whippersnappers, whipping boy

    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
     Infórmanos
    -0.63
    Hentet
    -0.53
    carpa
    -0.53
    featureID
    -0.51
    ellido
    -0.51
    PropertyChanging
    -0.50
    ardu
    -0.50
     Boyz
    -0.50
    initas
    -0.50
    Konrad
    -0.49
    POSITIVE LOGITS
     Whip
    0.85
     whip
    0.85
    Whip
    0.82
     whipped
    0.79
     whipping
    0.75
    whip
    0.75
     Whipple
    0.74
     Whi
    0.71
    Whi
    0.68
     whips
    0.66
    Activations Density 0.005%

    No Known Activations