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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 125753
    Prev
    Next
    INDEX
    Explanations

    serum

    np_max-act · gemini-2.0-flash

    The neuron detects numeric measurement tokens—especially concentration values and percentages (e.g. “10 %,” decimal quantities, etc.).

    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
    izioni
    -0.07
     nests
    -0.06
    agrid
    -0.06
     Suite
    -0.06
    üfus
    -0.06
    storeId
    -0.06
    .mob
    -0.06
    ึ้
    -0.06
     manifold
    -0.06
    getEmail
    -0.06
    POSITIVE LOGITS
     Tim
    0.07
    =".$
    0.07
     Katie
    0.06
    ะแ
    0.06
    ==$
    0.06
     =>
    0.06
    _taken
    0.06
    *.
    0.06
     मतलब
    0.06
    _ERRORS
    0.06
    Activations Density 0.000%

    No Known Activations