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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 14012
    Prev
    Next
    INDEX
    Explanations

    turbine

    np_max-act · gemini-2.0-flash

    The neuron activates on mention of distinct gas turbine engine components or sections (e.g. fan section, compressor section, combustor section, turbine section).

    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
    _fit
    -0.07
     Але
    -0.07
    _runtime
    -0.06
     XY
    -0.06
    /about
    -0.06
     runs
    -0.06
     TSR
    -0.06
    /R
    -0.06
     drinking
    -0.06
     hal
    -0.06
    POSITIVE LOGITS
    appoint
    0.08
     Smarty
    0.07
     vert
    0.07
     установ
    0.06
     gioc
    0.06
     Cavaliers
    0.06
     Pods
    0.06
    cano
    0.06
    aphael
    0.06
     gathered
    0.06
    Activations Density 0.004%

    No Known Activations