© 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 (Instruct)
    4. Resid Post - 131k
    5. 11-RESID-POST-AA
    6. 44191
    Prev
    Next
    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    structured data formats and configurations.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron is primarily detecting occurrences of the “@” character (at‐sign) in 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
    .nano
    -0.07
     uno
    -0.07
    ESP
    -0.07
     раньше
    -0.06
    _inactive
    -0.06
    新的
    -0.06
     simpler
    -0.06
    ankan
    -0.06
    hp
    -0.06
    ühl
    -0.06
    POSITIVE LOGITS
    าน
    0.06
     Collection
    0.06
    �
    0.06
     région
    0.06
    azines
    0.06
    ETS
    0.06
    ocard
    0.06
    —
    0.06
    et
    0.05
     xlink
    0.05
    Activations Density 0.077%

    No Known Activations