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

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2026
    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. 89994
    Prev
    Next
    INDEX
    Explanations

    Law and politics

    np_max-act · gemini-2.0-flash

    This neuron fires on domain‐specific entity names—that is, specialized nouns or titles (e.g. “Chief Justice,” “Supreme Court,” “12-mm scope,” “long double,” “real,” “imaginary”) that denote formal offices, technical instruments, units, or data types.

    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
    (Device
    -0.08
    <link
    -0.07
    averse
    -0.06
     Kil
    -0.06
     BMC
    -0.06
    _BINARY
    -0.06
    Payload
    -0.06
    .firstname
    -0.06
     accru
    -0.06
    _TH
    -0.06
    POSITIVE LOGITS
    ू
    0.07
     boots
    0.07
     kes
    0.06
    iffe
    0.06
     ним
    0.06
    FK
    0.06
     "\""
    0.06
     misguided
    0.06
    PATCH
    0.06
     있던
    0.06
    Activations Density 0.170%

    No Known Activations