Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 109508
    Prev
    Next
    INDEX
    Explanations

    scientific articles

    np_max-act · gemini-2.0-flash

    The neuron is triggered by mentions of Gram‐positive bacteria (e.g. “gram‐positive bacterium Bacillus subtilis”).

    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
     jihad
    -0.06
    continue
    -0.06
    ());↵↵↵
    -0.06
    Drop
    -0.06
    حاد
    -0.06
    qry
    -0.06
     Fame
    -0.06
    ening
    -0.06
    _deleted
    -0.06
    로
    -0.06
    POSITIVE LOGITS
     SCE
    0.06
     mycket
    0.06
     feudal
    0.06
     uranium
    0.06
     RTL
    0.06
    ��
    0.06
     mcc
    0.06
    umbai
    0.06
    <label
    0.06
    oid
    0.06
    Activations Density 0.055%

    No Known Activations