© 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. 1-RES-SM
    11. 28026
    Prev
    Next
    INDEX
    Explanations

    instances of the word "make" and its derivatives in various contexts

    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/1-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.1.hook_resid_post
    Hook Layer
    1
    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
    53
    +0.13
    0.7%
    407
    +0.12
    0.6%
    485
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    407
    +0.13
    0.07
    485
    +0.12
    0.07
    1
    +0.11
    0.06
    Negative Logits
    ¡
    -1.94
    nier
    -1.82
    ¥
    -1.76
    Ń
    -1.75
    enthal
    -1.70
    ¤
    -1.68
    mma
    -1.68
    omitempty
    -1.63
    gi
    -1.61
    ij
    -1.60
    POSITIVE LOGITS
    hift
    2.17
     sure
    1.94
     merry
    1.74
    leine
    1.73
    peace
    1.67
     haste
    1.48
     periodic
    1.46
     matters
    1.44
     preparations
    1.44
     AMERICA
    1.43
    Activations Density 0.189%

    No Known Activations