© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 13887
    Prev
    Next
    INDEX
    Explanations

    it was

    np_acts-logits-general · gemini-2.5-flash-lite

    Expressions of counterfactual “if not for” or “had it not been for” conditional clauses.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     hardly
    0.56
    たる
    0.56
    establishing
    0.54
    ুর
    0.52
    ôn
    0.52
    пред
    0.52
    ITTING
    0.52
    멈
    0.52
     réfléch
    0.51
     ư
    0.51
    POSITIVE LOGITS
     Wasn
    0.75
     manuss
    0.58
     wasn
    0.58
    был
    0.56
     weren
    0.55
    就这样
    0.54
    ᴥ
    0.54
     Eastern
    0.54
    じゃない
    0.54
    ɭ
    0.53
    Activations Density 0.033%

    No Known Activations