© 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. 14949
    Prev
    Next
    INDEX
    Explanations

    C++ syntax and function definitions

    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
    271
    +0.17
    1.0%
    156
    +0.15
    0.9%
    50
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    283
    +0.17
    0.04
    131
    +0.15
    0.04
    50
    +0.13
    0.04
    Negative Logits
    ited
    -1.75
    ycin
    -1.68
    uated
    -1.57
    ptin
    -1.56
    mered
    -1.55
    fitting
    -1.55
    itures
    -1.54
     thereof
    -1.51
    azole
    -1.49
    atically
    -1.49
    POSITIVE LOGITS
    ¿
    2.49
    ¡
    2.41
    »
    2.39
    Ļª
    2.31
    ¥
    2.24
    ¤
    2.20
    ¸
    2.17
    Ĵ
    2.16
    ½
    2.15
    Ĥ
    2.13
    Activations Density 0.175%

    No Known Activations