© 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. 24114
    Prev
    Next
    INDEX
    Explanations

    technical specifications and numerical values associated with computer hardware

    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/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
    Not in Any Lists

    No Comments

    Neuron Alignment
    Index
    Value
    % of L₁
    458
    +0.16
    0.9%
    362
    +0.15
    0.8%
    290
    +0.14
    0.8%
    Correlated Neurons
    Index
    P. Corr.
    Cos Sim.
    458
    +0.16
    0.06
    290
    +0.15
    0.06
    307
    +0.14
    0.04
    Negative Logits
     whole
    -1.74
    ese
    -1.52
    \]]{}
    -1.48
    ncbi
    -1.45
    apers
    -1.45
    akers
    -1.44
    dependencies
    -1.42
    oles
    -1.42
    uck
    -1.40
    dAtA
    -1.37
    POSITIVE LOGITS
     ($(
    1.67
     tracks
    1.57
    umns
    1.52
    ÅĽÄĩ
    1.49
    teenth
    1.43
    naire
    1.43
     ($
    1.39
    cad
    1.39
    th
    1.39
    tery
    1.38
    Activations Density 0.177%

    No Known Activations