© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW 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. 14987
    Prev
    Next
    INDEX
    Explanations

    references to color and its attributes

    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
    269
    +0.14
    0.8%
    9
    +0.14
    0.8%
    116
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    116
    +0.14
    0.03
    193
    +0.14
    0.03
    283
    +0.12
    0.03
    Negative Logits
    enix
    -2.07
    áĢ¯
    -2.00
    áĢ¬
    -1.99
    overnment
    -1.86
    ean
    -1.84
    aeda
    -1.83
    áŁ
    -1.73
    glich
    -1.70
    ETHOD
    -1.69
    gian
    -1.59
    POSITIVE LOGITS
    imetric
    2.66
    blind
    2.58
     palette
    2.58
    ado
    2.35
    ful
    2.26
    imetry
    2.20
    picker
    2.07
     scheme
    2.05
     gam
    2.04
    less
    2.03
    Activations Density 0.224%

    No Known Activations