© 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. 10914
    Prev
    Next
    INDEX
    Explanations

    references to kits, specifically in the context of kitchen or organizational tools

    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
    397
    +0.14
    0.8%
    87
    +0.11
    0.6%
    173
    +0.11
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    173
    +0.14
    0.01
    66
    +0.11
    0.02
    112
    +0.11
    0.01
    Negative Logits
    Ļª
    -2.35
    ppler
    -1.67
     time
    -1.67
     distant
    -1.64
     previous
    -1.51
    esides
    -1.45
    idences
    -1.44
    ateral
    -1.44
    shore
    -1.43
    ieurs
    -1.43
    POSITIVE LOGITS
    chen
    2.55
    ty
    2.44
    ters
    1.93
    neys
    1.77
    Į
    1.76
    ches
    1.73
    aphys
    1.71
    emat
    1.61
    ics
    1.60
    icle
    1.59
    Activations Density 0.181%

    No Known Activations