© 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. 24539
    Prev
    Next
    INDEX
    Explanations

    instances of the term "Dun" in the text

    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
    410
    +0.18
    1.0%
    369
    +0.16
    0.9%
    338
    +0.13
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    410
    +0.18
    0.02
    338
    +0.16
    0.02
    59
    +0.13
    0.02
    Negative Logits
     fixing
    -1.75
     responding
    -1.65
    PED
    -1.61
    acting
    -1.60
     communicating
    -1.58
    ogn
    -1.55
    Į
    -1.54
    EVER
    -1.50
     carrying
    -1.46
    ample
    -1.45
    POSITIVE LOGITS
    ettes
    2.19
    geons
    2.00
    ietz
    1.92
    ford
    1.73
    hurst
    1.71
    nett
    1.70
    geon
    1.70
    ette
    1.67
    ia
    1.66
    heim
    1.65
    Activations Density 0.120%

    No Known Activations