© 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. 9764
    Prev
    Next
    INDEX
    Explanations

    instances of indication or warning

    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
    466
    +0.18
    1.1%
    348
    +0.17
    1.0%
    307
    +0.17
    1.0%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    180
    +0.18
    0.04
    97
    +0.17
    0.05
    50
    +0.17
    0.05
    Negative Logits
    oslav
    -1.56
    urop
    -1.54
    jä
    -1.51
    iske
    -1.48
    olars
    -1.45
    oltz
    -1.45
    ontal
    -1.44
    ão
    -1.43
    zerba
    -1.40
    aments
    -1.39
    POSITIVE LOGITS
    whether
    2.11
    namely
    1.97
     whatever
    1.92
    especially
    1.92
    even
    1.75
    particularly
    1.72
    perhaps
    1.66
     whether
    1.65
    because
    1.64
    specifically
    1.61
    Activations Density 0.430%

    No Known Activations