© 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. 14394
    Prev
    Next
    INDEX
    Explanations

    references to research methods and their implementation in scientific studies

    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
    459
    +0.14
    0.8%
    208
    +0.13
    0.7%
    12
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    330
    +0.14
    0.06
    208
    +0.13
    0.03
    476
    +0.10
    0.05
    Negative Logits
     etc
    -1.94
    etc
    -1.88
    !”
    -1.87
     whatsoever
    -1.73
     aren
    -1.73
     EVERY
    -1.58
    thems
    -1.54
    !_
    -1.54
    laughter
    -1.52
    ”?
    -1.51
    POSITIVE LOGITS
    ĥ½
    2.13
    °
    2.06
    ł
    2.00
    ¯
    1.96
    ij
    1.94
     Briefly
    1.91
     proposed
    1.89
    Ħ
    1.89
     proposes
    1.88
     METHOD
    1.88
    Activations Density 0.580%

    No Known Activations