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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-2B
    5. 22-GEMMASCOPE-TRANSCODER-16K
    7. 12299
    Prev
    Next
    INDEX
    Explanations

    descriptions of physical altercations, particularly in a prison setting

    oai_token-act-pair · gemini-2.0-flash

    physical pushing/throwing actions

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2b-pt-transcoders/layer_22/width_16k/average_l0_15
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    16,384
    Data Type
    float32
    Hook Name
    blocks.22.ln2.hook_normalized
    Architecture
    jumprelu_transcoder
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ც
    -0.49
     fre
    -0.42
    peur
    -0.40
     FRE
    -0.39
     conduct
    -0.39
    nave
    -0.39
    nelle
    -0.39
    gesetzt
    -0.38
    fif
    -0.38
    กล
    -0.38
    POSITIVE LOGITS
     Pushing
    1.12
     pushing
    1.11
     throwing
    1.08
    Pushing
    1.06
     throw
    1.04
     threw
    0.98
     pushed
    0.98
     hitting
    0.98
     pushes
    0.97
     Throwing
    0.95
    Activations Density 2.875%

    No Known Activations