© 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. 12445
    Prev
    Next
    INDEX
    Explanations

    textual references to legal cases and decisions

    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
    78
    +0.18
    1.0%
    306
    +0.14
    0.8%
    506
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    78
    +0.18
    -0.00
    254
    +0.14
    0.00
    506
    +0.10
    0.00
    Negative Logits
     Usage
    -1.69
    SEM
    -1.41
    References
    -1.31
    CRIPTION
    -1.31
    matically
    -1.31
    mium
    -1.30
    abol
    -1.27
    mentation
    -1.25
     concentrate
    -1.23
    ctions
    -1.23
    POSITIVE LOGITS
    teenth
    1.95
    th
    1.85
    eenth
    1.79
    ermine
    1.72
    teen
    1.67
    alia
    1.58
    umb
    1.57
    arynx
    1.55
    ona
    1.54
    rums
    1.50
    Activations Density 0.065%

    No Known Activations