© 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. 2-RES-SM
    11. 31781
    Prev
    Next
    INDEX
    Explanations

    phrases that indicate approximate quantities or measurements

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    110
    +0.13
    0.7%
    86
    +0.12
    0.7%
    246
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    485
    +0.13
    0.02
    110
    +0.12
    0.01
    102
    +0.11
    0.02
    Negative Logits
    over
    -1.61
    ename
    -1.60
    ocker
    -1.56
    ney
    -1.55
    neys
    -1.55
    anda
    -1.52
    être
    -1.52
     administrator
    -1.50
    arin
    -1.47
    ";
    -1.45
    POSITIVE LOGITS
     aligned
    1.66
     neighb
    1.64
     expected
    1.64
    coming
    1.58
     doubled
    1.57
     doubling
    1.56
    GREEK
    1.51
     70
    1.50
     sized
    1.48
     average
    1.47
    Activations Density 0.091%

    No Known Activations