© 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. 13559
    Prev
    Next
    INDEX
    Explanations

    keywords and terminology related to programming libraries and directives

    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
    357
    +0.15
    0.9%
    348
    +0.15
    0.8%
    29
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    341
    +0.15
    0.02
    78
    +0.15
    0.00
    357
    +0.14
    0.03
    Negative Logits
    lessly
    -1.54
    eted
    -1.50
    iers
    -1.49
    arily
    -1.44
    heres
    -1.44
     degree
    -1.44
     amended
    -1.43
    akh
    -1.43
    ounces
    -1.42
    etically
    -1.41
    POSITIVE LOGITS
    ¼
    7.98
    °
    7.87
    Ń
    7.81
    ª
    7.75
    §
    7.71
    ¸
    7.69
    ½
    7.69
    Ĭ
    7.66
    ·
    7.65
    ¯
    7.56
    Activations Density 0.114%

    No Known Activations