© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    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. 2-RES-SM
    11. 12484
    Prev
    Next
    INDEX
    Explanations

    technical or programming-related syntax elements and characters

    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/2-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.2.hook_resid_post
    Hook Layer
    2
    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
    301
    +0.13
    0.8%
    40
    +0.13
    0.7%
    155
    +0.11
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    301
    +0.13
    0.07
    46
    +0.13
    0.03
    408
    +0.11
    0.08
    Negative Logits
    Ļª
    -1.73
    aders
    -1.67
    ĨĴ
    -1.63
    ı
    -1.57
    Ĥ¬
    -1.57
     &
    -1.48
    $]{};
    -1.43
    º
    -1.43
    »
    -1.41
    .+
    -1.41
    POSITIVE LOGITS
    Argued
    1.43
     ATCC
    1.42
     inventor
    1.41
    later
    1.35
    versed
    1.35
     offices
    1.34
    otted
    1.34
     helm
    1.33
    atinib
    1.33
    funded
    1.32
    Activations Density 4.059%

    No Known Activations