© 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. 29513
    Prev
    Next
    INDEX
    Explanations

    references to silicon and related materials in the context of technology and manufacturing

    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
    156
    +0.26
    1.6%
    71
    +0.13
    0.8%
    152
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    71
    +0.26
    0.02
    275
    +0.13
    0.01
    446
    +0.12
    0.01
    Negative Logits
     furious
    -1.51
      ↵
    -1.51
     wait
    -1.50
    lers
    -1.49
     sob
    -1.47
    matically
    -1.44
     gentle
    -1.43
     mercy
    -1.43
     stopping
    -1.41
     sleep
    -1.40
    POSITIVE LOGITS
    ."[
    1.89
     dioxide
    1.80
    µ
    1.64
    anium
    1.63
    inux
    1.62
    ®
    1.61
    anguage
    1.61
    arium
    1.60
    istor
    1.55
    ieux
    1.54
    Activations Density 0.055%

    No Known Activations