© 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
    2. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    3. Llama3.1-8B-IT
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 125003
    Prev
    Next
    INDEX
    Explanations

    see

    np_max-act · gemini-2.0-flash

    The neuron detects occurrences of “see” (as in “see …”) in comments, marking cross-references to documentation or other code.

    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
     measure
    -0.07
    �
    -0.07
     quarters
    -0.07
     груд
    -0.06
            ↵        ↵        ↵
    -0.06
    มน
    -0.06
     off
    -0.06
    !I
    -0.06
    IDTH
    -0.06
     fö
    -0.06
    POSITIVE LOGITS
    \Product
    0.07
    ану
    0.06
    _firestore
    0.06
    asury
    0.06
    แทน
    0.06
    _staff
    0.06
    .Process
    0.06
    opsis
    0.06
    ्टर
    0.06
    uso
    0.06
    Activations Density 0.015%

    No Known Activations