© 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. 126812
    Prev
    Next
    INDEX
    Explanations

    climbing trees

    np_max-act · gemini-2.0-flash

    This neuron never activates on any meaningful token—it’s effectively “dead” and doesn’t respond to anything.

    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
    SHORT
    -0.08
    .mas
    -0.07
    _circle
    -0.07
    库
    -0.06
     overseas
    -0.06
    .alt
    -0.06
     =================================================
    -0.06
    _stop
    -0.06
    _hour
    -0.06
    omas
    -0.06
    POSITIVE LOGITS
     wsp
    0.06
    로그램
    0.06
     secular
    0.06
    +w
    0.06
    xEB
    0.06
    ,)↵
    0.06
    reur
    0.06
    Can
    0.06
    riet
    0.06
    firefox
    0.06
    Activations Density 0.029%

    No Known Activations