© 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
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 73817
    Prev
    Next
    INDEX
    Explanations

    classes

    np_max-act · gemini-2.0-flash

    This neuron specifically fires on instances of the “classes” label or identifier (e.g. the “All Classes” navigation link/ID) in API documentation HTML.

    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
     })
    -0.06
     Provider
    -0.06
     Detection
    -0.06
    .AF
    -0.06
     Wood
    -0.06
     &=
    -0.06
    уже
    -0.06
    ाख
    -0.06
    .MouseDown
    -0.05
     Armour
    -0.05
    POSITIVE LOGITS
    READING
    0.08
    MC
    0.07
    lingen
    0.07
     팬
    0.07
    ULA
    0.06
    _non
    0.06
     instructed
    0.06
     wasted
    0.06
    某
    0.06
    CLUDING
    0.06
    Activations Density 0.000%

    No Known Activations