© 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. 7909
    Prev
    Next
    INDEX
    Explanations

    code snippets

    np_max-act · gemini-2.0-flash

    The neuron is detecting indent or leading‐whitespace tokens (i.e. runs of spaces or tabs used for code indentation).

    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
    zim
    -0.06
     Dahl
    -0.06
    .ep
    -0.06
     malé
    -0.06
    мон
    -0.06
    柄
    -0.06
     اعتر
    -0.06
     вим
    -0.06
     jejichž
    -0.06
    cke
    -0.06
    POSITIVE LOGITS
    лен
    0.08
    /Test
    0.07
    ='<
    0.07
    LOC
    0.06
    _Process
    0.06
    (low
    0.06
     أمريكي
    0.06
     img
    0.06
    <_
    0.06
    Bag
    0.06
    Activations Density 0.000%

    No Known Activations