© 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. 2-RES-SM
    11. 27530
    Prev
    Next
    INDEX
    Explanations

    elements of a structured data format or programming-related content

    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
    56
    +0.14
    0.8%
    263
    +0.13
    0.8%
    340
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    408
    +0.14
    0.09
    105
    +0.13
    0.09
    219
    +0.11
    0.08
    Negative Logits
    brand
    -1.46
     constants
    -1.38
    idation
    -1.37
    uffix
    -1.37
     value
    -1.36
    lies
    -1.36
     label
    -1.36
     precedent
    -1.35
    orandum
    -1.35
     quantity
    -1.34
    POSITIVE LOGITS
    Ĺ
    1.79
    ½
    1.78
    ¥
    1.77
    ¿½
    1.66
    İ
    1.62
    ŀ
    1.61
    ¤
    1.52
    ĻĤ
    1.52
    ³
    1.50
    ĵ
    1.48
    Activations Density 6.099%

    No Known Activations