© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 91522
    Prev
    Next
    INDEX
    Explanations

    sensitivity

    np_max-act · gemini-2.0-flash

    The neuron consistently flags subword tokens that are part of clinical measurement names or biomarker terms (e.g. “sensitivity,” “gly‐,” “hs” in “hs‐CRP,” etc.).

    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
    ازم
    -0.07
     Investigators
    -0.07
     مشار
    -0.07
     quantidade
    -0.06
    _handle
    -0.06
    Posts
    -0.06
     amateur
    -0.06
    mime
    -0.06
    icont
    -0.06
    .LA
    -0.06
    POSITIVE LOGITS
     isAuthenticated
    0.06
     autoload
    0.06
    (PDO
    0.06
     MatDialog
    0.06
    дах
    0.06
     privateKey
    0.06
    aginator
    0.06
    (jq
    0.06
    DUCT
    0.06
    _chance
    0.06
    Activations Density 0.005%

    No Known Activations