© 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. 4-RES-SM
    11. 14314
    Prev
    Next
    INDEX
    Explanations

    terms related to quantum physics

    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
    47
    +0.13
    0.7%
    422
    +0.12
    0.7%
    2
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    422
    +0.13
    0.02
    2
    +0.12
    0.01
    346
    +0.12
    0.01
    Negative Logits
    ership
    -2.06
    ercase
    -1.77
    ETHOD
    -1.73
    oser
    -1.65
    illion
    -1.57
    oted
    -1.57
    hest
    -1.56
    igh
    -1.55
    oting
    -1.53
    ards
    -1.50
    POSITIVE LOGITS
    ¿
    2.05
    ĵ
    2.00
    ¥
    1.96
    ī
    1.92
     electro
    1.85
     mechanically
    1.77
     mechanics
    1.75
    ķ
    1.73
    ¡
    1.71
    fusion
    1.71
    Activations Density 0.131%

    No Known Activations