Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 27224
    Prev
    Next
    INDEX
    Explanations

    demands

    np_max-act · gemini-2.0-flash

    This neuron detects language used for legal demands or cease-and-desist style threats.

    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
    BeforeEach
    -0.07
    论
    -0.07
    ३
    -0.06
    vection
    -0.06
    _grp
    -0.06
    xFD
    -0.06
    Chip
    -0.06
     Palo
    -0.06
    рес
    -0.06
     dependable
    -0.06
    POSITIVE LOGITS
     xe
    0.07
    !
    ↵
    0.06
    ayah
    0.06
    .setOnItemClickListener
    0.06
     reps
    0.06
     Kişisel
    0.06
     Enhanced
    0.06
    .Deserialize
    0.06
     FormsModule
    0.06
     paramInt
    0.06
    Activations Density 0.026%

    No Known Activations