© 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. 31971
    Prev
    Next
    INDEX
    Explanations

    prepositions and articles

    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
    23
    +0.16
    0.9%
    478
    +0.15
    0.8%
    176
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    404
    +0.16
    0.04
    238
    +0.15
    0.03
    176
    +0.11
    0.04
    Negative Logits
    ī
    -1.93
    stown
    -1.78
    Īĺ
    -1.77
    zent
    -1.71
    haus
    -1.71
    enthal
    -1.61
    point
    -1.53
    ħ
    -1.52
    eur
    -1.51
    ij
    -1.48
    POSITIVE LOGITS
     remission
    1.82
     significant
    1.76
    elig
    1.71
    completion
    1.59
    significant
    1.57
    umerable
    1.57
     lasting
    1.55
     disappearance
    1.55
     translocation
    1.54
     relapse
    1.53
    Activations Density 0.264%

    No Known Activations