© 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. 125138
    Prev
    Next
    INDEX
    Explanations

    el

    np_max-act · gemini-2.0-flash

    The neuron specifically detects mentions of the disease “multiple myeloma,” especially when its name is tokenized (e.g., “my-,” “el-,” “oma”).

    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
     Berk
    -0.07
    sequence
    -0.06
     Paran
    -0.06
    俊
    -0.06
     опред
    -0.06
    LoginForm
    -0.06
     K
    -0.06
    chk
    -0.06
     огранич
    -0.06
    indrome
    -0.06
    POSITIVE LOGITS
    (AL
    0.07
     noble
    0.07
    APA
    0.07
    ]);
    0.06
    _IA
    0.06
     objectMapper
    0.06
     Smoke
    0.06
     AA
    0.06
    RAM
    0.06
     LOWER
    0.06
    Activations Density 0.004%

    No Known Activations