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. 43951
    Prev
    Next
    INDEX
    Explanations

    colon

    np_max-act · gemini-2.0-flash

    The neuron is specialized to pick out colon‐delimited hexadecimal address segments (e.g. the 2-digit hex pairs found in MAC or IPv6 addresses).

    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
    Chars
    -0.07
     ******************************************************************************↵
    -0.07
    .userID
    -0.06
    nx
    -0.06
    SharedPreferences
    -0.06
    .transactions
    -0.06
     StatusCode
    -0.06
    //!↵
    -0.06
     qint
    -0.06
    शक
    -0.06
    POSITIVE LOGITS
    각
    0.06
     strate
    0.06
     vál
    0.06
     eğit
    0.06
     المغ
    0.06
     RL
    0.06
    osit
    0.06
    spor
    0.06
    ltra
    0.06
     ру
    0.06
    Activations Density 0.004%

    No Known Activations