© 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. 90910
    Prev
    Next
    INDEX
    Explanations

    cv2 code

    np_max-act · gemini-2.0-flash

    The neuron never activates on any content tokens—it appears to be entirely inactive (a “dead” neuron).

    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
    stoupil
    -0.07
    Mas
    -0.07
    ????????
    -0.07
    beeld
    -0.07
     nossa
    -0.07
     impres
    -0.07
    .offsetHeight
    -0.06
    bed
    -0.06
    SharedPtr
    -0.06
    营业
    -0.06
    POSITIVE LOGITS
    .waitKey
    0.10
     Lists
    0.07
    MENTS
    0.07
     reusable
    0.06
     realms
    0.06
     bulls
    0.06
    KEY
    0.06
     LIST
    0.06
    �
    0.06
    aney
    0.06
    Activations Density 0.000%

    No Known Activations