© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI 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. 68470
    Prev
    Next
    INDEX
    Explanations

    Intelligence stat

    np_max-act · gemini-2.0-flash

    This neuron primarily activates on numeric tokens—especially D&D–style stat numbers and decimal values.

    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
    PCR
    -0.07
    符合
    -0.06
    班
    -0.06
     ("
    -0.06
     Compared
    -0.06
    Sur
    -0.06
    .Formatter
    -0.06
     pagamento
    -0.06
    办
    -0.06
    _LESS
    -0.06
    POSITIVE LOGITS
     Gew
    0.07
     skype
    0.06
    ục
    0.06
    studio
    0.06
    544
    0.06
    awy
    0.06
    ﾆﾆﾆﾆ
    0.06
    mass
    0.06
     دلیل
    0.06
     immer
    0.06
    Activations Density 0.003%

    No Known Activations