© 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
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 55990
    Prev
    Next
    INDEX
    Explanations

    code diffs

    np_max-act · gemini-2.0-flash

    The neuron is detecting version‐control “diff” metadata markers (e.g. the “a/… b/…,” “---,” “+++,” and related diff‐header tokens).

    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
     Cli
    -0.08
    ुत
    -0.07
     MessageType
    -0.06
    じゃない
    -0.06
    rowsable
    -0.06
    _windows
    -0.06
    _password
    -0.06
    communications
    -0.06
    fontSize
    -0.06
    材料
    -0.06
    POSITIVE LOGITS
    how
    0.08
     Authorities
    0.07
     aload
    0.07
    坡
    0.06
    끔
    0.06
     Called
    0.06
     гром
    0.06
    اء
    0.06
     уп
    0.06
     aaa
    0.06
    Activations Density 0.001%

    No Known Activations