© 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. 10632
    Prev
    Next
    INDEX
    Explanations

    references to financial metrics and valuations

    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
    415
    +0.17
    1.0%
    342
    +0.12
    0.7%
    369
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    415
    +0.17
    0.04
    342
    +0.12
    0.04
    369
    +0.11
    0.03
    Negative Logits
    views
    -1.49
    fulness
    -1.42
    googleapis
    -1.39
     feed
    -1.38
    zzi
    -1.37
    ction
    -1.35
    paper
    -1.35
    stalk
    -1.32
    son
    -1.31
    hood
    -1.29
    POSITIVE LOGITS
    ĻĤ
    3.12
    ĨĴ
    2.69
    Ļª
    2.61
    ı
    2.58
    inetics
    2.54
    ĩ
    2.49
    ij
    2.49
    orean
    2.44
    Ģ
    2.40
    Ľ
    2.38
    Activations Density 0.094%

    No Known Activations