© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 57197
    Prev
    Next
    INDEX
    Explanations

    invalid

    np_max-act · gemini-2.0-flash

    This neuron detects occurrences of the substring “invalid” as part of identifiers or error codes.

    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
     Chore
    -0.07
     Monter
    -0.07
     orchest
    -0.07
    .Tree
    -0.06
     gor
    -0.06
    puter
    -0.06
     recurrence
    -0.06
     recur
    -0.06
     tú
    -0.06
     komt
    -0.06
    POSITIVE LOGITS
     invalid
    0.11
    invalid
    0.10
     Invalid
    0.10
    Invalid
    0.09
    .invalid
    0.08
    -invalid
    0.08
    ни
    0.08
     NV
    0.08
    HUD
    0.07
     Evans
    0.07
    Activations Density 0.004%

    No Known Activations