© 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. 2-RES-SM
    11. 30316
    Prev
    Next
    INDEX
    Explanations

    conjunctions and expressives that emphasize connection or continuation

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    478
    +0.15
    0.9%
    34
    +0.12
    0.7%
    50
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    494
    +0.15
    0.05
    288
    +0.12
    0.08
    50
    +0.11
    0.05
    Negative Logits
     {#
    -1.90
     versus
    -1.75
     depends
    -1.67
     information
    -1.60
    ior
    -1.52
    ness
    -1.48
     (“
    -1.47
    ÅĽci
    -1.43
    ison
    -1.43
    information
    -1.39
    POSITIVE LOGITS
     subsequent
    1.70
    ahoma
    1.61
    days
    1.61
     later
    1.50
     onwards
    1.49
    era
    1.44
     onward
    1.44
     hence
    1.42
     eventual
    1.40
     culmin
    1.38
    Activations Density 0.547%

    No Known Activations