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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Google DeepMind · Exploring Gemma 2 with Gemma Scope
    5. Gemma-2-2B
    7. Residual Stream - 16k
    9. 18-GEMMASCOPE-RES-16K
    11. 7050
    Prev
    Next
    INDEX
    Explanations

    phrases related to law enforcement or legal duties

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

    Following "of," "my," or "to" followed by "job" or "duty"

    np_acts-logits-general · gemini-2.0-flash

    possessive pronoun + job or duty

    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
     שוליים
    -0.69
    Geplaatst
    -0.63
    ReusableCell
    -0.56
    Composable
    -0.53
    rungsseite
    -0.51
    Palmar
    -0.48
    aneity
    -0.48
    enroll
    -0.47
     estimés
    -0.46
    PageNumber
    -0.46
    POSITIVE LOGITS
     job
    1.77
     duties
    1.55
    job
    1.47
     Job
    1.43
    Job
    1.42
     duty
    1.38
     JOB
    1.37
     jobs
    1.34
     Duties
    1.27
    JOB
    1.26
    Activations Density 0.276%

    No Known Activations