© 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. 109799
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    The neuron activates on words that express malicious or harmful sentiment (e.g. “hurtful,” “malevolent,” “harm,” “venomous”).

    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
    _PLUS
    -0.07
     Ô
    -0.07
    Ê
    -0.07
    โปรแกรม
    -0.07
     Mongolia
    -0.07
     Monad
    -0.07
     status
    -0.06
    andum
    -0.06
    CellStyle
    -0.06
     Statistics
    -0.06
    POSITIVE LOGITS
    _bl
    0.07
    Find
    0.06
     disposal
    0.06
     Tear
    0.06
     hood
    0.06
     Name
    0.06
     outbound
    0.06
     interpolation
    0.06
     br
    0.06
     tent
    0.06
    Activations Density 0.005%

    No Known Activations