© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 89039
    Prev
    Next
    INDEX
    Explanations

    second

    np_max-act · gemini-2.0-flash

    The neuron activates on ordinal labels and step indicators (e.g. “Stage 2,” “second loop”) that mark numbered steps or phases in procedural descriptions.

    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
    _the
    -0.07
    usr
    -0.07
    _REQUIRED
    -0.06
    .fromRGBO
    -0.06
    edriver
    -0.06
    ตำ
    -0.06
    oth
    -0.06
    .Enter
    -0.06
    _rg
    -0.06
    =\'
    -0.06
    POSITIVE LOGITS
    volent
    0.07
     Про
    0.06
    كيب
    0.06
    indsight
    0.06
    ocial
    0.06
    fulWidget
    0.06
    INESS
    0.06
    .assertIn
    0.06
    AYOUT
    0.06
     TRE
    0.06
    Activations Density 0.038%

    No Known Activations