© 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. 109337
    Prev
    Next
    INDEX
    Explanations

    Rearranging letters to form words

    np_max-act · gemini-2.0-flash

    This neuron detects occurrences of the word “letter” (especially “letters”) and discussions of character‐level frequency.

    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
     ^^
    -0.06
    @Transactional
    -0.06
    芬
    -0.06
    초
    -0.06
     NIGHT
    -0.06
     jenis
    -0.06
     comrades
    -0.05
    esion
    -0.05
     =↵
    -0.05
     CHANNEL
    -0.05
    POSITIVE LOGITS
    among
    0.07
     benefiting
    0.07
    .toDouble
    0.07
    ataires
    0.07
    rough
    0.06
    bar
    0.06
     jsonify
    0.06
    .phi
    0.06
    [__
    0.06
     ได
    0.06
    Activations Density 0.020%

    No Known Activations