© 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. 30598
    Prev
    Next
    INDEX
    Explanations

    image morphology

    np_max-act · gemini-2.0-flash

    This neuron detects mentions of morphological image‐processing operations (e.g., “morphology,” “morphological,” “connected‐components,” 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
     cog
    -0.07
     Cow
    -0.07
    _resume
    -0.06
    -switch
    -0.06
    precation
    -0.06
     estudio
    -0.06
    _priority
    -0.06
     Tor
    -0.06
     Mec
    -0.06
     hog
    -0.06
    POSITIVE LOGITS
    ifiable
    0.07
     thức
    0.07
     SESSION
    0.07
    знача
    0.07
     triple
    0.06
     ordered
    0.06
    ової
    0.06
     nå
    0.06
    &M
    0.06
     studied
    0.06
    Activations Density 0.004%

    No Known Activations