© 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. 41241
    Prev
    Next
    INDEX
    Explanations

    product releases

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of new product releases or scheduling details (e.g. “new,” dates, release times, and related announcement terms).

    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
     value
    -0.07
     infamous
    -0.06
     acknowledgement
    -0.06
    력이
    -0.06
    素
    -0.06
     dragon
    -0.06
     solved
    -0.06
     contradiction
    -0.06
    _es
    -0.06
     brackets
    -0.06
    POSITIVE LOGITS
     Hawai
    0.06
     tưởng
    0.06
     swirling
    0.06
    MethodImpl
    0.06
     můj
    0.06
     전세
    0.06
    [keys
    0.06
     VIN
    0.06
    šk
    0.06
    juries
    0.06
    Activations Density 0.008%

    No Known Activations