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. 86114
    Prev
    Next
    INDEX
    Explanations

    feasible

    np_max-act · gemini-2.0-flash

    The neuron spotlights evaluative terms and modal constructions (like “can,” “feasible,” “effective,” “useful,” “integrated,” etc.) that assert a method’s or intervention’s feasibility or effectiveness.

    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
     Strikes
    -0.07
    .Option
    -0.07
     reviews
    -0.06
     DOCUMENT
    -0.06
    Yaw
    -0.06
    ifie
    -0.06
     confessed
    -0.06
    ),"
    -0.06
     результ
    -0.06
    Coeff
    -0.06
    POSITIVE LOGITS
    (inst
    0.06
    -return
    0.06
    .addListener
    0.06
     đổ
    0.06
     Norse
    0.06
    -metal
    0.06
     nghiên
    0.06
    _past
    0.06
    -girl
    0.06
     programmed
    0.06
    Activations Density 0.051%

    No Known Activations