© 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. 14433
    Prev
    Next
    INDEX
    Explanations

    references to copyright information and legal notices

    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
    79
    +0.12
    0.7%
    388
    +0.12
    0.7%
    263
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    388
    +0.12
    0.02
    425
    +0.12
    0.05
    202
    +0.12
    0.00
    Negative Logits
    PEC
    -1.45
     comprising
    -1.44
     comprises
    -1.42
    pshire
    -1.40
    PLC
    -1.38
     comprise
    -1.36
    )}/\
    -1.35
    '):
    -1.34
     semicon
    -1.32
    EXPORT
    -1.30
    POSITIVE LOGITS
     Instagram
    1.34
     ©
    1.32
     Berlin
    1.28
    zen
    1.28
     Facebook
    1.27
    ights
    1.27
     followers
    1.24
    ounces
    1.24
    ÑĥÑģ
    1.24
     Story
    1.24
    Activations Density 0.382%

    No Known Activations