© 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. 109682
    Prev
    Next
    INDEX
    Explanations

    Research papers

    np_max-act · gemini-2.0-flash

    The neuron flags dense, multiword scientific‐ or technical‐jargon expressions (e.g. compound method names, measurement terms, and specialized chemical or process phrases).

    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
    /products
    -0.07
     selecting
    -0.07
    ocale
    -0.06
    �
    -0.06
     Seller
    -0.06
    cut
    -0.06
     servicios
    -0.06
    ค
    -0.06
    -radio
    -0.06
    cell
    -0.06
    POSITIVE LOGITS
    乎
    0.07
     //--------------------------------
    0.06
    े,
    0.06
    陶
    0.06
    ват
    0.06
     росій
    0.06
     "\">
    0.06
    术
    0.06
    -Д
    0.06
     [#
    0.06
    Activations Density 0.152%

    No Known Activations