© 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. 126031
    Prev
    Next
    INDEX
    Explanations

    model

    np_max-act · gemini-2.0-flash

    This neuron responds to uses of the Keras functional API’s Model class (i.e. occurrences of “Model” imported from `keras.models`).

    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
     messages
    -0.08
     WhatsApp
    -0.07
     керів
    -0.07
    ají
    -0.06
     Zahl
    -0.06
     aster
    -0.06
    Dead
    -0.06
          
    -0.06
    vector
    -0.06
     Savior
    -0.06
    POSITIVE LOGITS
     현
    0.08
    [name
    0.07
    ایط
    0.07
     Thames
    0.07
     emerges
    0.06
     tin
    0.06
    )::
    0.06
     enters
    0.06
    thought
    0.06
    FRING
    0.06
    Activations Density 0.004%

    No Known Activations