© 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. 13163
    Prev
    Next
    INDEX
    Explanations
    No Explanations Found
    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
    188
    +0.14
    0.8%
    355
    +0.12
    0.6%
    232
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    0
    -0.14
    0.00
    1
    -0.12
    0.00
    2
    -0.11
    0.00
    Negative Logits
    ĥ½
    -1.61
    olip
    -1.56
    ateur
    -1.54
    /âĪĴ
    -1.49
    ocytes
    -1.49
     fibres
    -1.49
    ĥ
    -1.46
    ¯
    -1.45
    nem
    -1.44
    Ģ
    -1.42
    POSITIVE LOGITS
    hte
    1.71
    pert
    1.62
     himself
    1.60
    hing
    1.60
     Himself
    1.57
    aeda
    1.56
    '):
    1.55
     disrespect
    1.51
    thon
    1.51
     Heights
    1.48
    Activations Density 0.000%

    No Known Activations

    This feature has no known activations.