© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 41535
    Prev
    Next
    INDEX
    Explanations

    atic

    np_max-act · gemini-2.0-flash

    The neuron fires on mentions of metastasis or metastatic disease.

    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
     executed
    -0.08
    Bell
    -0.07
     writeFile
    -0.07
    #----------------------------------------------------------------
    -0.06
     remained
    -0.06
    .Qual
    -0.06
    -----------↵
    -0.06
     pride
    -0.06
     endregion
    -0.06
     Bened
    -0.06
    POSITIVE LOGITS
    atic
    0.07
    .Fatal
    0.07
    argc
    0.07
    _FAR
    0.07
    ibil
    0.06
    OMIC
    0.06
     MOTOR
    0.06
     vac
    0.06
     mosaic
    0.06
    bbb
    0.06
    Activations Density 0.001%

    No Known Activations