© 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. 9220
    Prev
    Next
    INDEX
    Explanations

    references to individuals and their professional titles or roles

    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
    417
    +0.21
    1.2%
    145
    +0.17
    0.9%
    158
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    158
    +0.21
    0.04
    397
    +0.17
    0.03
    21
    +0.14
    0.02
    Negative Logits
    unov
    -1.95
    orov
    -1.72
    ÃŁ
    -1.51
     RD
    -1.49
    urs
    -1.38
     Ice
    -1.37
     LN
    -1.33
     walks
    -1.33
     handed
    -1.32
     VEGF
    -1.31
    POSITIVE LOGITS
     Com
    3.25
     com
    2.99
    Com
    2.84
     COM
    2.68
    COM
    2.67
    com
    2.55
     comet
    2.26
     combo
    2.20
     composite
    2.17
    comm
    2.05
    Activations Density 0.626%

    No Known Activations