© 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 (Instruct)
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 30575
    Prev
    Next
    INDEX
    Explanations

    -N

    np_max-act · gemini-2.0-flash

    This neuron responds to alphanumeric, hyphenated chemical or compound identifiers (e.g. “C-NETA,” “NE‐3TA,” “TRX‐PF2001”).

    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
     vyvol
    -0.07
     seg
    -0.07
    abric
    -0.07
    istency
    -0.07
     seus
    -0.06
     Thrones
    -0.06
     periodo
    -0.06
     EQUAL
    -0.06
     consistency
    -0.06
    _sq
    -0.06
    POSITIVE LOGITS
    ΩΤ
    0.06
    agento
    0.06
    Loc
    0.06
    CUSTOM
    0.06
    COVER
    0.06
    ुँ
    0.06
    (FLAGS
    0.06
    ूँ
    0.06
    ……。
    0.06
    INTERN
    0.06
    Activations Density 0.143%

    No Known Activations