© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Under Peer Review · Sparse Autoencoders for Pythia-70M-Deduped
    5. Pythia-70M-Deduped
    7. Residuals
    9. 2-RES-SM
    11. 26527
    Prev
    Next
    INDEX
    Explanations

    references to licenses

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    361
    +0.14
    0.8%
    111
    +0.13
    0.8%
    2
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    361
    +0.14
    0.02
    145
    +0.13
    0.01
    107
    +0.13
    0.01
    Negative Logits
    ĨĴ
    -2.75
    Īĺ
    -2.69
    ı
    -2.69
    Į
    -2.67
    ·
    -2.67
    ®
    -2.66
    Ģ
    -2.65
    -2.64
    ↵↵
    -2.64
    ↵↵
    -2.64
    POSITIVE LOGITS
     organizations
    1.71
     enterprises
    1.62
    ious
    1.62
     firms
    1.61
     Redist
    1.58
     companies
    1.51
     applications
    1.48
     businesses
    1.45
     municipalities
    1.44
     state
    1.43
    Activations Density 0.050%

    No Known Activations