© 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. 28388
    Prev
    Next
    INDEX
    Explanations

    the preposition "to" 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
    23
    +0.22
    1.3%
    69
    +0.20
    1.2%
    478
    +0.15
    0.9%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.22
    0.06
    505
    +0.20
    0.03
    20
    +0.15
    0.03
    Negative Logits
    dag
    -1.77
    "</
    -1.71
    piper
    -1.64
    ató
    -1.58
    ↵³³
    -1.56
     awarded
    -1.56
    listed
    -1.50
    kerchief
    -1.47
    over
    -1.46
     tribut
    -1.45
    POSITIVE LOGITS
    ¥
    1.70
     designing
    1.62
     various
    1.60
     our
    1.56
     either
    1.55
    ķ
    1.55
    verting
    1.51
    Ħ
    1.47
    ·
    1.47
     varying
    1.45
    Activations Density 0.266%

    No Known Activations