© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 4-RES-SM
    11. 10954
    Prev
    Next
    INDEX
    Explanations

    code-related patterns or structures in technical documentation

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/4-res-sm
    Prompts (Dashboard)
    32,768 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    Features
    32,768
    Data Type
    torch.float32
    Hook Name
    blocks.4.hook_resid_post
    Hook Layer
    4
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    21
    +0.25
    1.5%
    78
    +0.19
    1.1%
    320
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    320
    +0.25
    0.03
    353
    +0.19
    0.03
    144
    +0.13
    0.02
    Negative Logits
     fav
    -1.59
    ?**
    -1.55
     snaps
    -1.52
     mates
    -1.46
    LLY
    -1.45
     favour
    -1.43
     pedestrians
    -1.42
    rfloor
    -1.42
    gements
    -1.41
    »¿
    -1.39
    POSITIVE LOGITS
    using
    2.87
    namespace
    2.54
    foreach
    2.10
     namespace
    1.89
    };
    1.82
    static
    1.77
    typedef
    1.75
    extern
    1.69
     foreach
    1.68
    ///
    1.61
    Activations Density 0.082%

    No Known Activations