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. 86324
    Prev
    Next
    INDEX
    Explanations

    string splitting

    np_max-act · gemini-2.0-flash

    The neuron activates on mentions of splitting strings by a delimiter (e.g. calls to “split” and references to spaces, commas, tabs, or other separators).

    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
     dusty
    -0.08
    moduleId
    -0.07
     blocks
    -0.07
     đoàn
    -0.07
    _block
    -0.07
    .scroll
    -0.07
     Swords
    -0.06
    /native
    -0.06
    ъек
    -0.06
     creatures
    -0.06
    POSITIVE LOGITS
     PSA
    0.06
     geldi
    0.05
     getDate
    0.05
     simmer
    0.05
    pq
    0.05
    entieth
    0.05
     Note
    0.05
     Impro
    0.05
     manufacturers
    0.05
     gibi
    0.05
    Activations Density 0.015%

    No Known Activations