© 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. 73745
    Prev
    Next
    INDEX
    Explanations

    technical problems with inventions

    np_max-act · gemini-2.0-flash

    The neuron responds to terms that signal technical procedures or problem statements—words about making or improving something and noting difficulty or effort (e.g. “making,” “improvement,” “difficult”).

    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
    rops
    -0.07
    imations
    -0.06
     real
    -0.06
     toes
    -0.06
    gr
    -0.06
    omen
    -0.06
     furniture
    -0.06
    ould
    -0.06
    Transactional
    -0.06
    $PostalCodesNL
    -0.06
    POSITIVE LOGITS
     currentValue
    0.07
    _linked
    0.07
     fName
    0.07
     locker
    0.07
     ště
    0.07
     phố
    0.07
    FormField
    0.06
     Nonetheless
    0.06
     Nevertheless
    0.06
    InternalServerError
    0.06
    Activations Density 0.055%

    No Known Activations