Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    APICircuit TracerNEWSteerSAE EvalsExportsSlackBlogPrivacy & TermsContact
    © Neuronpedia 2025
    Privacy & TermsBlogGitHubSlackTwitterContact
    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. 130639
    Prev
    Next
    INDEX
    Explanations

    of

    np_max-act · gemini-2.0-flash

    discussions on experimental methodologies in scientific research.

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

    This neuron is effectively “dead”—it never activates on any input and thereby detects no feature.

    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
    пов
    -0.07
     Keith
    -0.06
    dependent
    -0.06
     อย
    -0.06
     Grove
    -0.06
     ремон
    -0.06
     Firewall
    -0.06
    xBE
    -0.06
    _compat
    -0.06
     دش
    -0.06
    POSITIVE LOGITS
    rud
    0.06
    [msg
    0.06
     unlock
    0.06
    <?>>
    0.06
    GP
    0.06
    .executeQuery
    0.06
     jusqu
    0.06
     *);↵↵
    0.06
    236
    0.06
    .series
    0.06
    Activations Density 0.029%

    No Known Activations