© 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. 9659
    Prev
    Next
    INDEX
    Explanations

    questions or phrases that inquire about processes or methods

    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
    307
    +0.28
    1.6%
    240
    +0.17
    1.0%
    361
    +0.13
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    240
    +0.28
    0.03
    361
    +0.17
    0.06
    67
    +0.13
    0.03
    Negative Logits
    heses
    -1.77
    ión
    -1.71
     whom
    -1.65
    reon
    -1.57
    ees
    -1.57
     example
    -1.50
    AndroidRuntime
    -1.46
    orship
    -1.45
    ESULT
    -1.44
     dÃŃa
    -1.43
    POSITIVE LOGITS
    soever
    2.74
    itzer
    2.37
     much
    2.11
     they
    2.04
    ls
    1.94
    ling
    1.83
    much
    1.82
     we
    1.76
     many
    1.71
     badly
    1.68
    Activations Density 0.153%

    No Known Activations