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. 18-GEMMASCOPE-RES-16K
    6. 3039
    Prev
    Next
    INDEX
    Explanations

    references to legality and visibility related to criminal activities

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

    Overt or explicit expressions

    np_acts-logits-general · gemini-2.0-flash

    open, explicit, obvious

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.18.hook_resid_post
    Hook Layer
    18
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    MigrationBuilder
    -0.82
     <>",
    -0.63
    AccessorTable
    -0.60
    InvalidProtocol
    -0.60
    Infinite
    -0.60
    XmlAccessorType
    -0.59
     Infinite
    -0.58
    CloseOperation
    -0.57
    fjspx
    -0.57
     PackageManager
    -0.56
    POSITIVE LOGITS
     openly
    1.80
     overt
    1.50
     explicit
    1.40
     overtly
    1.36
     explicitly
    1.35
     blatant
    1.33
     blatantly
    1.32
     obvious
    1.27
     explí
    1.23
    explicit
    1.21
    Activations Density 0.325%

    No Known Activations