© 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. 12548
    Prev
    Next
    INDEX
    Explanations

    quotation mark

    np_max-act · gemini-2.0-flash

    This neuron detects numeric (especially floating-point) tokens—i.e. the scale values (like “1.0283203125”, “2.103515625”, “x”, etc.) in the JSON image metadata.

    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
    fft
    -0.07
     Karlov
    -0.07
    cee
    -0.06
    our
    -0.06
     Gors
    -0.06
     Giul
    -0.06
    robat
    -0.06
    ière
    -0.06
    tı
    -0.06
    .activity
    -0.06
    POSITIVE LOGITS
    cling
    0.06
    Very
    0.06
    $image
    0.06
    ...',↵
    0.06
    看到
    0.06
     Semiconductor
    0.06
     crafted
    0.06
    (pkt
    0.06
     '>'
    0.06
     PMC
    0.06
    Activations Density 0.001%

    No Known Activations