© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 58636
    Prev
    Next
    INDEX
    Explanations

    finding

    np_max-act · gemini-2.0-flash

    This neuron detects action‐oriented verbs (like “collect,” “need,” “find,” “defeat,” etc.) that appear in step‐by‐step instructional 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
     undermined
    -0.07
     Dallas
    -0.07
     obstacles
    -0.06
    れた
    -0.06
    links
    -0.06
     dll
    -0.06
     moral
    -0.06
    _history
    -0.06
     gallery
    -0.06
    це
    -0.06
    POSITIVE LOGITS
    /crypto
    0.07
    σιμο
    0.07
    ueva
    0.06
     společ
    0.06
    ΑΓ
    0.06
    brain
    0.06
    ��
    0.06
     removable
    0.06
    times
    0.06
    "context
    0.06
    Activations Density 0.009%

    No Known Activations