© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI 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. 39078
    Prev
    Next
    INDEX
    Explanations

    Capitalized text

    np_max-act · gemini-2.0-flash

    This neuron responds to the very first uppercase letter(s) at the start of titles or section headings.

    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
     Rep
    -0.07
    }],
    -0.06
    elden
    -0.06
     một
    -0.06
     mf
    -0.06
     Heat
    -0.06
    -modules
    -0.06
     label
    -0.06
     consoles
    -0.06
     rejects
    -0.06
    POSITIVE LOGITS
    instanc
    0.07
     сна
    0.06
     Municip
    0.06
    žil
    0.06
    ektiv
    0.06
     STRING
    0.06
    :NSLocalizedString
    0.06
    tls
    0.06
    .onDestroy
    0.06
    _DECREF
    0.06
    Activations Density 0.043%

    No Known Activations