© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 125054
    Prev
    Next
    INDEX
    Explanations

    Water jets and streams

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of using water—particularly streams, hoses, rinsing, flushing, or other pressurized water actions.

    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
    .party
    -0.06
    -0.06
    eygamber
    -0.06
    리고
    -0.06
    BackColor
    -0.06
    573
    -0.06
     Tan
    -0.06
     заступ
    -0.06
     Sür
    -0.05
     دیگران
    -0.05
    POSITIVE LOGITS
    (properties
    0.07
     corporate
    0.07
     consisting
    0.07
    traits
    0.07
     face
    0.06
     untrue
    0.06
     snd
    0.06
     theoretical
    0.06
     ___
    0.06
    struct
    0.06
    Activations Density 0.013%

    No Known Activations