© 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. 73839
    Prev
    Next
    INDEX
    Explanations

    radio

    np_max-act · gemini-2.0-flash

    The neuron detects mentions of radio broadcasting (e.g. the word “radio,” radio program names, and related broadcast media contexts).

    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
     estable
    -0.07
    leich
    -0.06
     melt
    -0.06
    โดย
    -0.06
     最
    -0.06
    Browse
    -0.06
     España
    -0.06
     surfing
    -0.06
     обол
    -0.06
    lama
    -0.06
    POSITIVE LOGITS
     Radio
    0.09
    Radio
    0.09
     RADIO
    0.08
     radio
    0.08
    _limits
    0.07
    (KERN
    0.07
    RadioButton
    0.07
    抬
    0.07
    _unregister
    0.06
     oxygen
    0.06
    Activations Density 0.004%

    No Known Activations