© 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. 1-RES-SM
    11. 30020
    Prev
    Next
    INDEX
    Explanations

    references to lamination

    oai_token-act-pair · gpt-4o-mini
    Triggered by @bot
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    ctigges/pythia-70m-deduped__res-sm_processed/1-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.1.hook_resid_post
    Hook Layer
    1
    Architecture
    standard
    Context Size
    128
    Dataset
    EleutherAI/the_pile_deduplicated
    Activation Function
    relu
    Embeds
    IFrame
    Link
    Neuron Alignment
    Index
    Value
    % of L₁
    59
    +0.12
    0.7%
    444
    +0.12
    0.7%
    492
    +0.12
    0.7%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    59
    +0.12
    0.01
    477
    +0.12
    0.01
    2
    +0.12
    0.01
    Negative Logits
     times
    -1.81
    ftware
    -1.67
     slightest
    -1.66
     wonder
    -1.65
    $).
    -1.65
     handy
    -1.61
     peace
    -1.59
     course
    -1.57
     magnitude
    -1.56
     treasure
    -1.56
    POSITIVE LOGITS
    ents
    2.19
    estone
    2.04
    krit
    1.87
    eson
    1.82
    recht
    1.78
    ieux
    1.77
    ĻĤ
    1.77
    iano
    1.75
    inated
    1.75
    »
    1.75
    Activations Density 0.053%

    No Known Activations