© 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. 2-RES-SM
    11. 28809
    Prev
    Next
    INDEX
    Explanations

    references to digital technology and advancements

    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
    156
    +0.26
    1.6%
    10
    +0.14
    0.9%
    30
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    10
    +0.26
    0.02
    30
    +0.14
    0.02
    377
    +0.14
    0.02
    Negative Logits
    lla
    -1.50
     crick
    -1.49
     reasonable
    -1.48
     sudden
    -1.45
    orems
    -1.43
     unreasonable
    -1.41
    const
    -1.38
     drastic
    -1.37
     discipl
    -1.36
     solitary
    -1.34
    POSITIVE LOGITS
     versions
    1.81
     arts
    1.79
    irected
    1.74
    ios
    1.67
     counterparts
    1.63
    iom
    1.61
    caster
    1.61
     version
    1.57
    ieux
    1.57
     formats
    1.54
    Activations Density 0.068%

    No Known Activations