© 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
    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. 85703
    Prev
    Next
    INDEX
    Explanations

    percentages in scientific texts

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on numeric expressions and quantitative measurement tokens (e.g. numeric values, percentages, or statistical result figures).

    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
     Pikachu
    -0.08
     FA
    -0.07
     what
    -0.06
     INV
    -0.06
     improvised
    -0.06
    AYER
    -0.06
    ět
    -0.06
     PA
    -0.06
     consultancy
    -0.06
    PA
    -0.06
    POSITIVE LOGITS
     άλ
    0.07
    (NO
    0.06
     assort
    0.06
    �
    0.06
     işe
    0.06
    other
    0.06
    ラク
    0.06
     bulun
    0.06
     authDomain
    0.06
    Đối
    0.06
    Activations Density 0.076%

    No Known Activations