© 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 EvalsExportsAPI 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. 12432
    Prev
    Next
    INDEX
    Explanations

    medical terminology related to parasitic infections

    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
    +0.23
    1.3%
    28
    +0.11
    0.6%
    348
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    130
    +0.23
    0.05
    31
    +0.11
    0.04
    45
    +0.11
    0.03
    Negative Logits
    ĨĴ
    -2.14
    ij
    -1.99
    ł
    -1.99
    ı
    -1.94
    İ
    -1.84
    NESS
    -1.83
    -1.83
    ↵
    -1.83
                  ↵
    -1.83
    -1.83
    POSITIVE LOGITS
    ircraft
    2.44
    ppa
    2.39
    pp
    2.38
    pering
    2.28
    plasty
    2.17
    ires
    2.16
    per
    2.16
    resis
    2.00
    ea
    1.99
    ud
    1.98
    Activations Density 0.428%

    No Known Activations