© 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
    3. Gemma-3-1B
    5. 7-GEMMASCOPE-2-RES-16K
    7. 487
    Prev
    Next
    INDEX
    Explanations

    fraud and violation testing

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-1b-pt/resid_post/layer_7_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     읽
    1.32
    迭代
    1.28
     entwickeln
    1.24
     kroz
    1.21
     wygod
    1.19
     갖
    1.17
     međ
    1.16
    께
    1.15
     schaffen
    1.15
     그래프
    1.15
    POSITIVE LOGITS
     perpetrators
    1.83
     looting
    1.72
     perpetrated
    1.68
     atrocities
    1.60
     heinous
    1.59
     perpetrator
    1.57
     assaults
    1.57
     crimes
    1.57
     assault
    1.56
     criminals
    1.53
    Activations Density 3.711%

    No Known Activations