© 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. 41571
    Prev
    Next
    INDEX
    Explanations

    faithful/fidelity

    np_max-act · gemini-2.0-flash

    This neuron selectively activates on occurrences of “faithful”/“faithfulness” or “fidelity” (and their variants).

    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
    _ARR
    -0.07
     blocked
    -0.07
    236
    -0.07
     eliminate
    -0.07
    린이
    -0.07
     Zhang
    -0.06
    nop
    -0.06
     Rochester
    -0.06
     Dawn
    -0.06
     zwarte
    -0.06
    POSITIVE LOGITS
     fidelity
    0.12
     faithful
    0.09
     faithfully
    0.09
    idelity
    0.08
     fid
    0.08
     goodness
    0.07
     familiarity
    0.07
    IEL
    0.07
     getToken
    0.07
     isEqual
    0.07
    Activations Density 0.005%

    No Known Activations