© 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. 6816
    Prev
    Next
    INDEX
    Explanations

    to

    np_max-act · gemini-2.0-flash

    This neuron selectively fires on floating-point number tokens (e.g. “1.0390625”, “2.751953125”)—i.e. decimal numerals.

    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
    OTE
    -0.07
    Rgb
    -0.07
     beige
    -0.07
    oog
    -0.07
    GenericType
    -0.06
    	e
    -0.06
    laughs
    -0.06
    indices
    -0.06
    .SetKeyName
    -0.06
     Village
    -0.06
    POSITIVE LOGITS
     under
    0.07
     unchecked
    0.07
     imposes
    0.07
    öffent
    0.07
     @
    0.07
    ulg
    0.07
     applying
    0.07
    var
    0.06
    _PRED
    0.06
    atoria
    0.06
    Activations Density 0.020%

    No Known Activations