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. 1067
    Prev
    Next
    INDEX
    Explanations

    cold temperatures

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of specific temperature measurements (e.g. numeric values with “below,” “degrees,” or temperature units).

    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
    иж
    -0.07
    hv
    -0.07
     trees
    -0.07
    getModel
    -0.07
     отправ
    -0.06
    _mk
    -0.06
    .*(
    -0.06
    King
    -0.06
    County
    -0.06
    とき
    -0.06
    POSITIVE LOGITS
     στους
    0.07
     pneum
    0.06
     ніж
    0.06
     лицо
    0.06
     dış
    0.06
     Paste
    0.06
     Poke
    0.06
     відпов
    0.06
     specialized
    0.06
    (WIN
    0.06
    Activations Density 0.009%

    No Known Activations