© 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. 29119
    Prev
    Next
    INDEX
    Explanations

    bracketed expressions and parameters related to coding or programming contexts

    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
    470
    +0.13
    0.7%
    473
    +0.12
    0.6%
    478
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    69
    +0.13
    0.05
    478
    +0.12
    -0.01
    190
    +0.11
    0.04
    Negative Logits
    reng
    -1.58
    aines
    -1.42
    ascus
    -1.36
    burn
    -1.36
    ampa
    -1.34
    raw
    -1.32
    pero
    -1.31
    banks
    -1.25
    ubs
    -1.25
    ould
    -1.24
    POSITIVE LOGITS
     alone
    1.70
    ientes
    1.63
     academ
    1.54
     imperfect
    1.48
    ±
    1.45
     context
    1.43
     spite
    1.39
    giving
    1.39
    ductive
    1.39
    eding
    1.37
    Activations Density 0.856%

    No Known Activations