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

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & 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. 130709
    Prev
    Next
    INDEX
    Explanations

    investment timing

    np_max-act · gemini-2.0-flash

    This neuron activates on concrete investment‐advice action phrases (e.g. “buy,” “hold,” “stay invested,” “rebalance”) in a buy-and-hold strategy context.

    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
    atown
    -0.07
     Solic
    -0.06
     дис
    -0.06
     stati
    -0.06
    同意
    -0.06
    ænd
    -0.06
     Horde
    -0.06
     Vend
    -0.06
     tej
    -0.06
     برج
    -0.06
    POSITIVE LOGITS
     available
    0.07
    derive
    0.07
    خي
    0.06
    _props
    0.06
    available
    0.06
    .ACTION
    0.06
     درون
    0.06
    ška
    0.06
     advantages
    0.06
    线
    0.06
    Activations Density 0.010%

    No Known Activations