© 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. 10037
    Prev
    Next
    INDEX
    Explanations

    instances of the word "the."

    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
    21
    +0.20
    1.2%
    68
    +0.11
    0.7%
    454
    +0.10
    0.6%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    68
    +0.20
    0.02
    454
    +0.11
    0.01
    179
    +0.10
    0.02
    Negative Logits
    "}](#
    -1.45
    ize
    -1.41
    ctomy
    -1.41
     pumps
    -1.35
    ori
    -1.34
    ally
    -1.31
     visitors
    -1.30
    ries
    -1.30
    oge
    -1.28
     evoked
    -1.27
    POSITIVE LOGITS
     GNU
    1.96
    ľ
    1.77
     COPYRIGHT
    1.64
     GPL
    1.61
    ¸
    1.59
     WARRANTY
    1.59
    ·
    1.56
    IJ
    1.54
    ĥ
    1.52
    Ľ
    1.51
    Activations Density 0.044%

    No Known Activations