© 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. 24-GEMMASCOPE-RES-16K
    11. 14473
    Prev
    Next
    INDEX
    Explanations

    complex narratives surrounding moral dilemmas and criminal behavior

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

    associated with evil

    np_acts-logits-general · gemini-2.0-flash

    wickedness and evil acts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k
    Configuration
    google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73
    Prompts (Dashboard)
    36,864 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.24.hook_resid_post
    Hook Layer
    24
    Architecture
    jumprelu
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     refugi
    -0.63
     soñ
    -0.57
     hadas
    -0.54
    unanje
    -0.53
    miot
    -0.53
     disparu
    -0.52
    lių
    -0.52
     Format
    -0.52
     escase
    -0.51
    walde
    -0.51
    POSITIVE LOGITS
     evil
    1.08
     vile
    1.04
     vicious
    0.99
     diabo
    0.96
     wicked
    0.95
     cruel
    0.95
     depra
    0.94
     sinister
    0.93
     nef
    0.91
     despicable
    0.89
    Activations Density 0.509%

    No Known Activations