© 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. Gemma-2-9B-IT
    5. 20-AXBENCH-REFT-R1-RES-16K
    7. 153
    Prev
    Next
    INDEX
    Model
    gemma-2-9b-it
    Layer #
    20
    Steering Hook
    blocks.20.hook_resid_pre
    Steering Strength
    72.5
    Uploader
    bot-neuronpedia
    Created At
    2/15/2025 1:06:43 AM
    Raw Vector
    Actions
    Explanations

    repeated phrases or patterns, particularly emphasizing terms like "aka" or "wt"

    oai_token-act-pair · gpt-4o-mini
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    pyvene/gemma-reft-r1-9b-it-res/l20
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    httphttps
    -0.71
     esternos
    -0.67
    OGND
    -0.66
    $_['
    -0.66
    orsese
    -0.65
    setViewportView
    -0.63
    Personendaten
    -0.63
    estacks
    -0.62
    StructEnd
    -0.61
     nahilalakip
    -0.60
    POSITIVE LOGITS
    !
    0.38
    let
    0.36
     playfully
    0.36
    am
    0.36
    action
    0.34
     fun
    0.32
    ab
    0.32
    int
    0.31
     space
    0.31
    ka
    0.30
    Activations Density 0.002%

    No Known Activations