© 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. 6513
    Prev
    Next
    INDEX
    Explanations

    news articles

    np_max-act · gemini-2.0-flash

    The neuron detects author‐byline phrases—places where a reporter’s name is followed by a reporting verb (e.g. “Jonathan Law finds,” “Jack Evans sees,” “Barry Barber examines”).

    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
     opponents
    -0.08
    groups
    -0.07
    (IM
    -0.06
    kok
    -0.06
    suffix
    -0.06
    �
    -0.06
    appropriate
    -0.06
     ownership
    -0.06
    Forest
    -0.06
    ्व
    -0.06
    POSITIVE LOGITS
    DataGridView
    0.07
     nhé
    0.06
    žití
    0.06
    bbing
    0.06
    imizde
    0.06
    ccion
    0.06
    Transformation
    0.06
    (sess
    0.06
    Topology
    0.06
    áj
    0.06
    Activations Density 0.057%

    No Known Activations