Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APIAssistant AxisNEWCircuit TracerNEWSteerSAE EvalsExports Community BlogPrivacy & TermsContact
    © Neuronpedia 2025
    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. 91508
    Prev
    Next
    INDEX
    Explanations

    wavelength

    np_max-act · gemini-2.0-flash

    The neuron is detecting decimal number tokens (particularly floating‐point values, as in measured wavelengths).

    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
     Horror
    -0.08
     Stre
    -0.07
    (ic
    -0.07
     Refriger
    -0.07
    (stdin
    -0.07
    Info
    -0.07
     pare
    -0.07
    .Reverse
    -0.06
    "My
    -0.06
     obrov
    -0.06
    POSITIVE LOGITS
     wavelength
    0.13
     wavelengths
    0.12
    avelength
    0.09
     Å
    0.07
    �
    0.07
     вит
    0.07
     timestep
    0.07
    lambda
    0.07
    WP
    0.06
    tparam
    0.06
    Activations Density 0.003%

    No Known Activations