© 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. 24825
    Prev
    Next
    INDEX
    Explanations

    references to the concept of reality

    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
    110
    +0.13
    0.7%
    118
    +0.13
    0.7%
    66
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    116
    +0.13
    0.02
    66
    +0.13
    0.02
    352
    +0.12
    0.02
    Negative Logits
     personally
    -1.62
    CHANTABILITY
    -1.55
    Å¡ÃŃ
    -1.53
     senior
    -1.50
    ĻĤ
    -1.49
    >'
    -1.43
     parsley
    -1.43
    >\
    -1.42
    >';
    -1.41
    ammar
    -1.40
    POSITIVE LOGITS
    finder
    1.81
    iest
    1.76
     TV
    1.74
    movie
    1.71
     television
    1.67
    ography
    1.67
    IFIC
    1.64
    tele
    1.63
    ousel
    1.63
    making
    1.62
    Activations Density 0.118%

    No Known Activations