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. 84472
    Prev
    Next
    INDEX
    Explanations

    bind concat

    np_max-act · gemini-2.0-flash

    This neuron detects use of the pandas “concat” operation (e.g. calls to pd.concat) for combining DataFrames.

    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
    問
    -0.07
     Out
    -0.06
    _rel
    -0.06
     Gonzalez
    -0.06
     arter
    -0.06
    ember
    -0.06
    (Conv
    -0.06
    .marker
    -0.06
    /releases
    -0.06
     riding
    -0.06
    POSITIVE LOGITS
     techn
    0.07
    doctrine
    0.06
     prominent
    0.06
    τι
    0.06
     چهار
    0.06
     securities
    0.06
    $o
    0.06
    _tls
    0.06
     doctrines
    0.06
    .feature
    0.06
    Activations Density 0.007%

    No Known Activations