Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    1. Home
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 43663
    Prev
    Next
    INDEX
    Explanations

    script descriptions

    np_max-act · gemini-2.0-flash

    The neuron fires on the field names in a Homebrew “Formula” (e.g. desc, homepage, url, sha256, license, bottle, depends_on, 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
    rech
    -0.07
    ‚Ì
    -0.07
     Registration
    -0.07
    èo
    -0.07
    เศรษฐ
    -0.07
    �
    -0.06
    (ac
    -0.06
    .djangoproject
    -0.06
    (strategy
    -0.06
    _PLL
    -0.06
    POSITIVE LOGITS
     st
    0.07
     f
    0.06
     تمام
    0.06
     Austin
    0.06
     데
    0.06
     bara
    0.06
    IDEO
    0.06
     Ut
    0.06
     jt
    0.06
     dy
    0.06
    Activations Density 0.012%

    No Known Activations