© 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. 4-RES-SM
    11. 9581
    Prev
    Next
    INDEX
    Explanations

    references to numerical values, particularly relating to ranking or statistics

    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/4-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.4.hook_resid_post
    Hook Layer
    4
    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
    90
    +0.10
    0.6%
    232
    +0.10
    0.6%
    348
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    500
    +0.10
    0.04
    424
    +0.10
    0.03
    265
    +0.10
    0.01
    Negative Logits
    matically
    -1.55
    html
    -1.49
    rile
    -1.38
    gian
    -1.37
    uo
    -1.34
     Conduct
    -1.34
    medi
    -1.32
    raft
    -1.27
    aspx
    -1.26
     alia
    -1.23
    POSITIVE LOGITS
    ing
    2.88
     crunch
    2.07
    ishing
    1.77
    ality
    1.76
    ahead
    1.76
     of
    1.73
    ning
    1.70
    aire
    1.68
    less
    1.62
    borough
    1.60
    Activations Density 0.179%

    No Known Activations