© 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. 6773
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act · gemini-2.0-flash

    The neuron selectively activates on comparative/superlative constructions (e.g. “one of the largest,” “uma das melhores,” “one of the most…”).

    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
    .fm
    -0.06
    atorium
    -0.06
     την
    -0.06
    .conf
    -0.06
    bindung
    -0.06
    stüt
    -0.06
    атку
    -0.06
    ponce
    -0.06
     그의
    -0.06
     بيت
    -0.06
    POSITIVE LOGITS
    une
    0.08
     Edge
    0.07
    (DataType
    0.07
    .vx
    0.07
    (lr
    0.06
     :],
    0.06
     incred
    0.06
    _exe
    0.06
     Way
    0.06
    (dr
    0.06
    Activations Density 0.028%

    No Known Activations