© 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. 14892
    Prev
    Next
    INDEX
    Explanations

    patterns in code or configuration data structures

    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
    156
    +0.32
    1.9%
    78
    +0.19
    1.1%
    400
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    210
    +0.32
    0.04
    17
    +0.19
    0.06
    400
    +0.13
    0.03
    Negative Logits
    iciary
    -1.60
    ittal
    -1.51
    arily
    -1.50
    uary
    -1.49
    chemia
    -1.48
    ski
    -1.46
    mlin
    -1.44
    angular
    -1.43
    iti
    -1.42
    uates
    -1.38
    POSITIVE LOGITS
    ®
    5.01
    ¼
    4.92
    ĻĤ
    4.88
    ¬
    4.87
    ¶
    4.85
    Ĥ¬
    4.83
    Ĥ
    4.82
    ½
    4.79
    ¸
    4.72
    »
    4.66
    Activations Density 0.327%

    No Known Activations