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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 90578
    Prev
    Next
    INDEX
    Explanations

    ivial

    np_max-act · gemini-2.0-flash

    This neuron responds to mentions of commutativity, i.e. occurrences of “commutative” versus “non-commutative” in algebraic contexts.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    idity
    -0.06
    _Init
    -0.06
    acin
    -0.06
    MO
    -0.06
    azioni
    -0.06
    User
    -0.06
    vary
    -0.06
    Audio
    -0.06
     Bard
    -0.06
    izational
    -0.06
    POSITIVE LOGITS
     tjejer
    0.07
    場所
    0.07
     Spoon
    0.07
     تلاش
    0.06
     český
    0.06
     conjug
    0.06
     refreshToken
    0.06
     rewritten
    0.06
    (contract
    0.06
     contacting
    0.06
    Activations Density 0.008%

    No Known Activations