© 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. 14211
    Prev
    Next
    INDEX
    Explanations

    mathematical expressions related to common divisors or multiples

    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
    306
    +0.13
    0.7%
    275
    +0.12
    0.7%
    382
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    275
    +0.13
    0.01
    382
    +0.12
    0.01
    407
    +0.12
    0.01
    Negative Logits
    ary
    -1.87
    âĢŁ
    -1.83
    \'
    -1.79
    hops
    -1.75
    .\
    -1.67
    iment
    -1.59
    ual
    -1.54
    uality
    -1.53
    ometric
    -1.52
    ão
    -1.51
    POSITIVE LOGITS
    Ł
    2.90
    Īĺ
    2.88
    Ŀ
    2.83
    ħ
    2.73
    ŀ
    2.70
    ĥ
    2.64
    ĨĴ
    2.63
    ł
    2.59
    Ķ
    2.59
    ¡
    2.57
    Activations Density 0.023%

    No Known Activations