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. 21-GEMMASCOPE-RES-16K
    6. 1066
    Prev
    Next
    INDEX
    Explanations

    mentions of the abbreviation "UM" and variations of the word "hum."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 21-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_21/width_16k/average_l0_70
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.21.hook_resid_post
    Hook Layer
    21
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     makeStyles
    -0.56
    ArrowToggle
    -0.53
    chanel
    -0.53
    ⓧ
    -0.53
    nocześnie
    -0.51
     Paglinawan
    -0.51
    endregion
    -0.49
     Betracht
    -0.49
     getItemId
    -0.48
    setVerticalGroup
    -0.47
    POSITIVE LOGITS
     Bum
    0.69
    Bum
    0.67
    dum
    0.66
    Dum
    0.66
    UpInside
    0.65
     Dum
    0.63
     dum
    0.63
     DUM
    0.63
    gum
    0.60
    CSRF
    0.59
    Activations Density 0.100%

    No Known Activations