© 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
    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. 39080
    Prev
    Next
    INDEX
    Explanations

    code failures and errors

    np_max-act · gemini-2.0-flash

    The neuron fires on tokens that are part of error or exception messages (e.g. “Error”, “Exception”, “Assertion failed”, stack‐trace markers, 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
     αρχ
    -0.07
    ानसभ
    -0.07
     getWidth
    -0.07
    .Cell
    -0.07
    xing
    -0.07
    aque
    -0.06
    NUM
    -0.06
    _^(
    -0.06
     councillors
    -0.06
     здоров
    -0.06
    POSITIVE LOGITS
    ạch
    0.07
    خط
    0.06
    soever
    0.06
     heter
    0.06
     vastly
    0.06
     impending
    0.06
     Appropri
    0.06
     pretending
    0.06
     testing
    0.06
     Perfect
    0.06
    Activations Density 0.026%

    No Known Activations