© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 1016
    Prev
    Next
    INDEX
    Explanations

    Keycloak login/signup

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

    The neuron is primarily detecting apostrophes in words—that is, contractions and possessive forms.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    possessive constructions, particularly apostrophes indicating possession or contraction in technical and instructional contexts.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @jyhe0408

    punctuation-heavy technical formatting, especially apostrophes/quotes, hyphens and slashes, and standalone numbers, such as possessives, contractions, quoted UI labels, paths/endpoints, and codes.

    oai_token-act-pair · gpt-5Triggered 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
    Not in Any Lists

    No Comments

    Negative Logits
    programmed
    0.53
    EXE
    0.52
    void
    0.51
     संख
    0.49
    treated
    0.47
    shock
    0.47
    umping
    0.47
    proof
    0.47
    wedge
    0.47
    additive
    0.46
    POSITIVE LOGITS
     jwt
    0.88
     signUp
    0.87
     SignIn
    0.84
     signup
    0.84
     users
    0.81
     user
    0.80
     login
    0.80
    SignUp
    0.80
     SignUp
    0.80
     Login
    0.80
    Activations Density 0.003%

    No Known Activations