© 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
    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. 110316
    Prev
    Next
    INDEX
    Explanations

    reatment

    np_max-act · gemini-2.0-flash

    The neuron activates on terms describing experimental “pretreatment” or “treatment” steps (e.g., “Pretreatment of … with …”).

    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
    こそ
    -0.06
     smirk
    -0.06
    imate
    -0.06
     hare
    -0.06
    ेयर
    -0.06
     butto
    -0.06
    walk
    -0.06
     �
    -0.06
    ebin
    -0.06
    hek
    -0.06
    POSITIVE LOGITS
    .setToolTipText
    0.07
     LDS
    0.06
    ://{
    0.06
     seç
    0.06
    remely
    0.06
     dataGridViewCellStyle
    0.06
    localStorage
    0.06
    \xc
    0.06
     inexperienced
    0.06
     hungry
    0.06
    Activations Density 0.016%

    No Known Activations