© 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. 28703
    Prev
    Next
    INDEX
    Explanations

    references to military operations

    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
    301
    +0.14
    0.8%
    443
    +0.13
    0.7%
    509
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    322
    +0.14
    0.03
    301
    +0.13
    0.03
    443
    +0.11
    0.03
    Negative Logits
     '</
    -1.71
    what
    -1.66
     how
    -1.65
    lar
    -1.50
     what
    -1.49
     previous
    -1.45
     future
    -1.40
    doibase
    -1.39
     sucrose
    -1.39
    amm
    -1.39
    POSITIVE LOGITS
    ĨĴ
    2.38
    ¾
    1.96
    etable
    1.93
    enstein
    1.88
    edia
    1.84
     Arena
    1.74
    ede
    1.74
    breaker
    1.67
     Ground
    1.66
    ictionary
    1.65
    Activations Density 0.269%

    No Known Activations