© 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 (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 85756
    Prev
    Next
    INDEX
    Explanations

    damage and health

    np_max-act · gemini-2.0-flash

    The neuron detects terms describing parts or regions that are unaffected or on the opposite/remote side (e.g. “remote,” “uninjured,” “contralateral,” “unaffected”).

    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
    ya
    -0.06
     경기
    -0.06
     Allies
    -0.06
     progressing
    -0.06
    bery
    -0.06
    ifikace
    -0.06
     Catalyst
    -0.06
     piracy
    -0.06
    Ch
    -0.06
    _chr
    -0.05
    POSITIVE LOGITS
     unus
    0.07
     begs
    0.07
     kicker
    0.07
    ,[
    0.07
    919
    0.06
     سال
    0.06
     từ
    0.06
    (fun
    0.06
    ประเทศไทย
    0.06
     angered
    0.06
    Activations Density 0.043%

    No Known Activations