© 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. 10464
    Prev
    Next
    INDEX
    Explanations

    the word "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/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
    307
    +0.21
    1.2%
    231
    +0.17
    1.0%
    478
    +0.17
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    231
    +0.21
    0.02
    358
    +0.17
    0.02
    272
    +0.17
    0.02
    Negative Logits
    .);
    -1.83
    .\[
    -1.76
    ].)
    -1.71
    iary
    -1.69
    hors
    -1.66
    }\}$
    -1.61
     ensued
    -1.58
    \];
    -1.57
    felt
    -1.56
    \].
    -1.56
    POSITIVE LOGITS
    Ń
    2.76
    ķ
    2.61
    ĨĴ
    2.56
    ↵
    2.46
    ↵
    2.46
    ↵↵
    2.46
    <|outofrange|>
    2.46
              ↵
    2.46
       ↵
    2.46
     ↵
    2.46
    Activations Density 0.042%

    No Known Activations