© 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. 25030
    Prev
    Next
    INDEX
    Explanations

    occurrences of the word "spark."

    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
    115
    +0.13
    0.7%
    38
    +0.12
    0.7%
    321
    +0.12
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    38
    +0.13
    0.02
    321
    +0.12
    0.02
    74
    +0.12
    0.02
    Negative Logits
     behalf
    -1.81
    pping
    -1.72
    coming
    -1.66
    nae
    -1.58
    pped
    -1.56
    ĻĤ
    -1.50
    pload
    -1.49
    uthan
    -1.42
    opsies
    -1.41
    aeda
    -1.39
    POSITIVE LOGITS
    lers
    1.99
    ler
    1.79
    les
    1.76
    ley
    1.67
    alore
    1.65
    le
    1.65
    lings
    1.64
    uble
    1.64
    rock
    1.61
    eton
    1.60
    Activations Density 0.193%

    No Known Activations