© 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. 12652
    Prev
    Next
    INDEX
    Explanations

    hard coded

    np_max-act · gemini-2.0-flash

    This neuron activates on language referring to manual, fixed specifications—e.g. “hard-coded,” “hard-tuned,” or other mentions of explicitly setting fixed values.

    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
    padding
    -0.07
     wind
    -0.07
    Translate
    -0.07
    ë
    -0.07
    Kit
    -0.06
     sued
    -0.06
    通信
    -0.06
     robber
    -0.06
    ］
    -0.06
     Freud
    -0.06
    POSITIVE LOGITS
     reacted
    0.07
     �
    0.07
    /token
    0.06
    _String
    0.06
    [token
    0.06
    UFACT
    0.06
     Г
    0.06
     Shall
    0.06
     employ
    0.06
    。本
    0.06
    Activations Density 0.040%

    No Known Activations