© 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. 13611
    Prev
    Next
    INDEX
    Explanations

    mathematical terminology and expressions related to formal definitions and properties in a theoretical context

    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
    23
    +0.15
    0.9%
    125
    +0.14
    0.8%
    478
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    53
    +0.15
    0.06
    438
    +0.14
    0.05
    494
    +0.12
    0.03
    Negative Logits
    enders
    -2.15
    mog
    -1.81
    ncbi
    -1.75
     Awards
    -1.68
     featured
    -1.67
    creen
    -1.62
    estones
    -1.59
    entes
    -1.58
    endas
    -1.56
     submissions
    -1.55
    POSITIVE LOGITS
    ·¸
    2.88
    ¿½
    2.83
    Ī
    2.72
    ´
    2.71
    Ŀ
    2.55
    ©
    2.51
    ŀ
    2.50
    ľ
    2.44
    ¢
    2.42
    Ķ
    2.42
    Activations Density 0.606%

    No Known Activations