© 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. 1-RES-SM
    11. 28682
    Prev
    Next
    INDEX
    Explanations

    instances of the word "to," particularly in various forms and 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
    478
    +0.20
    1.2%
    69
    +0.19
    1.1%
    23
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.20
    0.10
    83
    +0.19
    0.06
    78
    +0.12
    0.05
    Negative Logits
    .+
    -1.74
    arity
    -1.67
    arily
    -1.67
    .-
    -1.62
    STM
    -1.62
    hole
    -1.55
    ENV
    -1.52
    ishly
    -1.51
    holes
    -1.50
    enthal
    -1.50
    POSITIVE LOGITS
     be
    1.80
    pless
    1.71
     rescue
    1.64
    oda
    1.58
     her
    1.57
    od
    1.53
     defend
    1.53
     investigate
    1.52
     favor
    1.51
     tackle
    1.50
    Activations Density 0.646%

    No Known Activations