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. 55286
    Prev
    Next
    INDEX
    Explanations

    Legal documents

    np_max-act · gemini-2.0-flash

    The neuron spotlights heavily indented or whitespace‐only tokens—i.e. lines that begin with large blocks of blank space.

    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
    美国
    -0.07
     murderer
    -0.07
     transparent
    -0.07
    deg
    -0.06
    َب
    -0.06
     strong
    -0.06
     NY
    -0.06
     Stones
    -0.06
     Arabs
    -0.06
     NZ
    -0.06
    POSITIVE LOGITS
    _BOOL
    0.06
     mír
    0.06
     voi
    0.06
    .failure
    0.06
    اير
    0.06
    ],[-
    0.06
    르는
    0.06
     уча
    0.06
    (parseInt
    0.06
    iveau
    0.06
    Activations Density 0.008%

    No Known Activations