© 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. 55382
    Prev
    Next
    INDEX
    Explanations

    bi

    np_max-act · gemini-2.0-flash

    This neuron fires on numeric tokens—especially floating‐point values—in code or API documentation (e.g. width/height parameter 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
    有的
    -0.06
    _Red
    -0.06
     Notes
    -0.06
    Envelope
    -0.06
    tor
    -0.06
    imoto
    -0.06
     bru
    -0.06
    .neo
    -0.06
     aff
    -0.06
     kami
    -0.06
    POSITIVE LOGITS
    ±n
    0.07
    ！
    0.07
    děl
    0.07
     Technique
    0.07
     тро
    0.06
    ̀
    0.06
    Factors
    0.06
    ACCEPT
    0.06
    uenta
    0.06
     usr
    0.06
    Activations Density 0.215%

    No Known Activations