© 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. 28434
    Prev
    Next
    INDEX
    Explanations

    references to various metals

    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
    +0.19
    1.1%
    275
    +0.11
    0.7%
    424
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    275
    +0.19
    0.02
    424
    +0.11
    0.01
    230
    +0.10
    0.01
    Negative Logits
     wait
    -2.03
     hers
    -1.85
     place
    -1.81
    TRODUCTION
    -1.67
     pause
    -1.62
     turn
    -1.57
     adopt
    -1.53
     duration
    -1.52
     yours
    -1.51
     another
    -1.50
    POSITIVE LOGITS
    anguage
    1.95
    lite
    1.92
    finger
    1.72
    finder
    1.71
    enium
    1.69
    scope
    1.67
    icas
    1.65
    theory
    1.62
    buntu
    1.61
     based
    1.60
    Activations Density 0.046%

    No Known Activations